fix: dtypes pickle format name

Dockerfile

@@ -5,7 +5,7 @@
 # tags from Docker Hub.
 FROM python:3.7-slim
 
-LABEL Name=dataset-converter Version=0.1.0
+LABEL Name=dataset-converter Version=0.1.1
 LABEL maintainer="Hugo Matalonga <[email protected]>"
 
 ARG UID=1000
@@ -21,8 +21,6 @@ RUN addgroup --system --gid ${GID} user \
 WORKDIR /home/user
 COPY ./app/requirements.txt /home/user
 
-# Using pip:
-RUN python3 -m pip install --upgrade --no-cache-dir --compile pip
 RUN python3 -m pip install --no-cache-dir --compile -r requirements.txt
 
 ADD ./entrypoint.sh /usr/local/bin
@@ -34,4 +32,4 @@ RUN chown -R user:user /home/user
 USER user
 ENV PATH=${PATH}:/home/user/.local/bin
 
-CMD ["/usr/local/bin/entrypoint.sh"]
+ENTRYPOINT ["/usr/local/bin/entrypoint.sh"]

README.md

@@ -28,6 +28,15 @@ $ docker-compose up
 ```
 - It will look for all `.yml`, for each dataset configured file, it will produce an optimized parquet file and a pickle file containing the pandas dtypes. The generated files are located in the `./data` folder.
 
+### Output files
+For each config file found, keeps the same file `name` as set in the config and create the following files:
+
+#### ${name}.dytpes.pickle
+Contains a dict python with the column:dtype for each entry.
+
+#### ${name}.parquet.7z
+Creates a parquet binary file compressed in 7z format from the dataframe processed.
+
 ## Plugins
 
 A plugin system is available, where is possible to call additional procedures to modify the dataset files.

app/app.py

@@ -72,7 +72,7 @@ def load_tasks(df, plugins, category):
 def export_files(df, name, compression):
   filename = name.split('.')[0]
 
-  filepath = os.path.join(data_path, filename + '.dtypes.p')
+  filepath = os.path.join(data_path, filename + '.dtypes.pickle')
   print('Creating dtypes file -> {}'.format(filepath))
   save_dtypes(cache_dtypes(df), filepath)