Stop hitting S3 in the smasher tests

.gitignore

@@ -33,6 +33,7 @@ workers/test_volume/*
 # However for a couple tests we do store the data in the repo and need to track them.
 !workers/test_volume/raw/TEST/TRANSCRIPTOME_INDEX/aegilops_tauschii_short.gtf.gz
 !workers/test_volume/raw/TEST/NO_OP/test.txt
+!workers/test_volume/cassettes
 
 
 # Byte-compiled / optimized / DLL files

workers/data_refinery_workers/processors/requirements.in

@@ -25,3 +25,5 @@ datashader>=0.6.9
 datashape>=0.5.2
 holoviews>=1.11.2
 selenium==3.141.0
+# For mocking out S3 calls for the smasher
+vcrpy

workers/data_refinery_workers/processors/requirements.txt

@@ -25,7 +25,7 @@ docopt==0.6.2             # via coveralls
 docutils==0.14            # via botocore
 heapdict==1.0.0           # via zict
 holoviews==1.11.3         # via -r requirements.in
-idna==2.7                 # via requests
+idna==2.7                 # via requests, yarl
 jinja2==2.11.1            # via -r requirements.in, bokeh
 jmespath==0.9.3           # via boto3, botocore
 joblib==0.13.2            # via scikit-learn
@@ -36,10 +36,11 @@ markupsafe==1.1.1         # via jinja2
 matplotlib==3.0.0         # via -r requirements.in, scikit-image
 mpmath==1.0.0             # via sympy
 msgpack==0.6.1            # via distributed
+multidict==4.7.6          # via yarl
 multipledispatch==0.6.0   # via datashape
 networkx==2.2             # via scikit-image
 numba==0.43.0             # via datashader
-numpy==1.15.2             # via -r requirements.in, bokeh, dask, datashader, datashape, holoviews, matplotlib, numba, pandas, pywavelets, scikit-learn, scipy, xarray
+numpy==1.15.2             # via -r requirements.in, bokeh, dask, datashader, datashape, holoviews, matplotlib, numba, pandas, pywavelets, scikit-learn, xarray
 packaging==19.0           # via bokeh
 pandas==0.23.4            # via -r requirements.in, dask, datashader, xarray
 param==1.8.2              # via colorcet, datashader, holoviews, pyct, pyviz-comms
@@ -54,7 +55,7 @@ python-nomad==1.1.0       # via -r requirements.in
 pytz==2018.5              # via django, pandas
 pyviz-comms==0.7.2        # via holoviews
 pywavelets==1.0.2         # via scikit-image
-pyyaml==4.2b4             # via -r requirements.in, bokeh, distributed, pyct
+pyyaml==4.2b4             # via -r requirements.in, bokeh, distributed, pyct, vcrpy
 requests==2.22.0          # via -r requirements.in, coveralls, pyct, python-nomad
 retrying==1.3.3           # via -r requirements.in
 s3transfer==0.2.1         # via boto3
@@ -63,7 +64,7 @@ scikit-learn==0.21.3      # via -r requirements.in
 scipy==1.1.0              # via -r requirements.in, datashader, scikit-image, scikit-learn
 selenium==3.141.0         # via -r requirements.in
 simplejson==3.16.0        # via -r requirements.in
-six==1.11.0               # via bokeh, cycler, distributed, multipledispatch, packaging, python-dateutil, retrying, scikit-image
+six==1.11.0               # via bokeh, cycler, distributed, multipledispatch, packaging, python-dateutil, retrying, scikit-image, vcrpy
 sortedcontainers==2.1.0   # via distributed
 sqlparse==0.3.0           # via django
 sympy==1.3                # via -r requirements.in
@@ -74,7 +75,10 @@ tornado==6.0.2            # via bokeh, distributed
 unicodecsv==0.14.1        # via -r requirements.in
 untangle==1.1.1           # via -r requirements.in
 urllib3==1.25.3           # via botocore, requests, selenium
+vcrpy==4.0.2              # via -r requirements.in
+wrapt==1.12.1             # via vcrpy
 xarray==0.12.3            # via -r requirements.in, datashader
+yarl==1.4.2               # via vcrpy
 zict==0.1.4               # via distributed
 
 # The following packages are considered to be unsafe in a requirements file:

workers/data_refinery_workers/processors/test_smasher.py

@@ -10,6 +10,8 @@
 from django.core.management import call_command
 from django.test import TransactionTestCase, tag
 
+import vcr
+
 from data_refinery_common.models import (
     ComputationalResult,
     ComputationalResultAnnotation,
@@ -1090,6 +1092,7 @@ def test_sanity_imports(self):
         import sympy  # noqa
 
     @tag("smasher")
+    @vcr.use_cassette("/home/user/data_store/cassettes/smasher.get_synced_files.yaml")
     def test_get_synced_files(self):
         """ """
         result = ComputationalResult()

workers/run_tests.sh

@@ -364,12 +364,10 @@ if [ -z "$tag" ] || [ "$tag" = "smasher" ] || [ "$tag" = "compendia" ]; then
         wget -q -O "$quant_test_data_1" \
              "$test_data_repo/$quant_name"
     fi
-    if [ -z "$AWS_ACCESS_KEY_ID" ]; then
-        AWS_ACCESS_KEY_ID="$(~/bin/aws configure get default.aws_access_key_id)"
-	export AWS_ACCESS_KEY_ID
-        AWS_SECRET_ACCESS_KEY="$(~/bin/aws configure get default.aws_secret_access_key)"
-	export AWS_SECRET_ACCESS_KEY
-    fi
+    # Mock out the AWS keys since we use VCR to mock out the request with these
+    # as the AWS credentials
+    export AWS_ACCESS_KEY_ID=XXX
+    export AWS_SECRET_ACCESS_KEY=XXX
 fi
 
 if [ -z "$tag" ] || [ "$tag" = "qn" ]; then