Update name of Pooler class to CSVPooler

tangledpath · Feb 24, 2024 · b1d24d2 · b1d24d2
1 parent 294f92f
commit b1d24d2
Show file tree

Hide file tree

Showing 8 changed files with 187 additions and 180 deletions.
diff --git a/README.md b/README.md
@@ -14,14 +14,14 @@ values:
 
 ### As dataframe row
 ```python
-  from csv_batcher.pooler import Pooler, CallbackWith
+  from csv_batcher.csv_pooler import CSVPooler, CallbackWith
 
   # Callback function passed to pooler; accepts a dataframe row
   #   as a pandas Series (via apply)
   def process_dataframe_row(row):
     return row.iloc[0]
 
-  pooler = Pooler(
+  pooler = CSVPooler(
     "5mSalesRecords.csv",
     process_dataframe_row,
     callback_with=CallbackWith.DATAFRAME_ROW,
@@ -31,7 +31,7 @@ values:
 
 ### As dataframe
 ```python
-  from csv_batcher.pooler import Pooler, CallbackWith
+  from csv_batcher.csv_pooler import CSVPooler, CallbackWith
 
   # Used in DataFrame.apply:
   def process_dataframe_row(row):
@@ -43,7 +43,7 @@ values:
     # Or do something more complicated....
     return len(df)
 
-  pooler = Pooler(
+  pooler = CSVPooler(
     "5mSalesRecords.csv",
     process_dataframe,
     callback_with=CallbackWith.DATAFRAME,
@@ -53,7 +53,7 @@ values:
 
 ### As CSV filename
 ```python
-  from csv_batcher.pooler import Pooler, CallbackWith
+  from csv_batcher.csv_pooler import CSVPooler, CallbackWith
 
   def process_csv_filename(csv_chunk_filename):
       # print("processing ", csv_chunk_filename)
@@ -68,7 +68,7 @@ values:
   def process_dataframe_row(row):
       return row.iloc[0]
 
-  pooler = Pooler(
+  pooler = CSVPooler(
     "5mSalesRecords.csv",
     process_dataframe,
     callback_with=CallbackWith.CSV_FILENAME
@@ -96,3 +96,6 @@ values:
 ```bash
   clear; pytest
 ```
+
+### Publishing
+`poetry publish --build -u __token__ -p $PYPI_TOKEN`
diff --git a/csv_batcher/csv_pooler.py b/csv_batcher/csv_pooler.py
@@ -21,13 +21,13 @@ class CallbackWith(StrEnum):
     #   function. This causes the callback function to be called with
     #   each row as a `pd.Series` object:
     DATAFRAME_ROW = auto()
-class Pooler:
+class CSVPooler:
     def __init__(
         self,
         csv_filename: str,
         process_fn: Callable,
         callback_with: CallbackWith = CallbackWith.CSV_FILENAME,
-        pool_size: int = 12,
+        pool_size: int = 5,
         chunk_lines: int = 10000,
     ):
         """

diff --git a/csv_batcher/test_csv_pooler.py b/csv_batcher/test_csv_pooler.py
@@ -1,6 +1,6 @@
 import unittest
 from csv_batcher.utils.time import time_and_log
-from csv_batcher.csv_pooler import Pooler, CallbackWith
+from csv_batcher.csv_pooler import CSVPooler, CallbackWith
 import pandas as pd
 
 def __process_dataframe_row(row):
@@ -16,17 +16,17 @@ def __process_as_dataframe(df):
 
 def test_big_file_as_csv():
     with time_and_log("test_big_file_as_csv"):
-        pooler = Pooler("5mSalesRecords.csv", __process_csv_filename)
+        pooler = CSVPooler("5mSalesRecords.csv", __process_csv_filename)
         pooler.process()
 
 def test_big_file_as_dataframe():
     with time_and_log("test_big_file_as_dataframe"):
-        pooler = Pooler("5mSalesRecords.csv", __process_as_dataframe, callback_with=CallbackWith.DATAFRAME)
+        pooler = CSVPooler("5mSalesRecords.csv", __process_as_dataframe, callback_with=CallbackWith.DATAFRAME)
         pooler.process()
 
 def test_big_file_as_dataframe_rows():
     with time_and_log("test_big_file_as_dataframe_rows"):
-        pooler = Pooler("5mSalesRecords.csv", __process_dataframe_row, callback_with=CallbackWith.DATAFRAME_ROW)
+        pooler = CSVPooler("5mSalesRecords.csv", __process_dataframe_row, callback_with=CallbackWith.DATAFRAME_ROW)
         pooler.process()
 
 def test_no_pooler():

diff --git a/docs/csv_batcher.html b/docs/csv_batcher.html
@@ -74,14 +74,14 @@ <h2 id="usage">Usage</h2>
 <h3 id="as-dataframe-row">As dataframe row</h3>
 
 <div class="pdoc-code codehilite">
-<pre><span></span><code>  <span class="kn">from</span> <span class="nn">csv_batcher.pooler</span> <span class="kn">import</span> <span class="n">Pooler</span><span class="p">,</span> <span class="n">CallbackWith</span>
+<pre><span></span><code>  <span class="kn">from</span> <span class="nn"><a href="csv_batcher/csv_pooler.html">csv_batcher.csv_pooler</a></span> <span class="kn">import</span> <span class="n">CSVPooler</span><span class="p">,</span> <span class="n">CallbackWith</span>
 
   <span class="c1"># Callback function passed to pooler; accepts a dataframe row</span>
   <span class="c1">#   as a pandas Series (via apply)</span>
   <span class="k">def</span> <span class="nf">process_dataframe_row</span><span class="p">(</span><span class="n">row</span><span class="p">):</span>
     <span class="k">return</span> <span class="n">row</span><span class="o">.</span><span class="n">iloc</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
 
-  <span class="n">pooler</span> <span class="o">=</span> <span class="n">Pooler</span><span class="p">(</span>
+  <span class="n">pooler</span> <span class="o">=</span> <span class="n">CSVPooler</span><span class="p">(</span>
     <span class="s2">&quot;5mSalesRecords.csv&quot;</span><span class="p">,</span>
     <span class="n">process_dataframe_row</span><span class="p">,</span>
     <span class="n">callback_with</span><span class="o">=</span><span class="n">CallbackWith</span><span class="o">.</span><span class="n">DATAFRAME_ROW</span><span class="p">,</span>
@@ -91,7 +91,7 @@ <h3 id="as-dataframe-row">As dataframe row</h3>
 
 <span class="c1">### As dataframe</span>
 <span class="err">```</span><span class="n">python</span>
-  <span class="kn">from</span> <span class="nn">csv_batcher.pooler</span> <span class="kn">import</span> <span class="n">Pooler</span><span class="p">,</span> <span class="n">CallbackWith</span>
+  <span class="kn">from</span> <span class="nn"><a href="csv_batcher/csv_pooler.html">csv_batcher.csv_pooler</a></span> <span class="kn">import</span> <span class="n">CSVPooler</span><span class="p">,</span> <span class="n">CallbackWith</span>
 
   <span class="c1"># Used in DataFrame.apply:</span>
   <span class="k">def</span> <span class="nf">process_dataframe_row</span><span class="p">(</span><span class="n">row</span><span class="p">):</span>
@@ -103,7 +103,7 @@ <h3 id="as-dataframe-row">As dataframe row</h3>
     <span class="c1"># Or do something more complicated....</span>
     <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="n">df</span><span class="p">)</span>
 
-  <span class="n">pooler</span> <span class="o">=</span> <span class="n">Pooler</span><span class="p">(</span>
+  <span class="n">pooler</span> <span class="o">=</span> <span class="n">CSVPooler</span><span class="p">(</span>
     <span class="s2">&quot;5mSalesRecords.csv&quot;</span><span class="p">,</span>
     <span class="n">process_dataframe</span><span class="p">,</span>
     <span class="n">callback_with</span><span class="o">=</span><span class="n">CallbackWith</span><span class="o">.</span><span class="n">DATAFRAME</span><span class="p">,</span>
@@ -113,7 +113,7 @@ <h3 id="as-dataframe-row">As dataframe row</h3>
 
 <span class="c1">### As CSV filename</span>
 <span class="err">```</span><span class="n">python</span>
-  <span class="kn">from</span> <span class="nn">csv_batcher.pooler</span> <span class="kn">import</span> <span class="n">Pooler</span><span class="p">,</span> <span class="n">CallbackWith</span>
+  <span class="kn">from</span> <span class="nn"><a href="csv_batcher/csv_pooler.html">csv_batcher.csv_pooler</a></span> <span class="kn">import</span> <span class="n">CSVPooler</span><span class="p">,</span> <span class="n">CallbackWith</span>
 
   <span class="k">def</span> <span class="nf">process_csv_filename</span><span class="p">(</span><span class="n">csv_chunk_filename</span><span class="p">):</span>
       <span class="c1"># print(&quot;processing &quot;, csv_chunk_filename)</span>
@@ -128,7 +128,7 @@ <h3 id="as-dataframe-row">As dataframe row</h3>
   <span class="k">def</span> <span class="nf">process_dataframe_row</span><span class="p">(</span><span class="n">row</span><span class="p">):</span>
       <span class="k">return</span> <span class="n">row</span><span class="o">.</span><span class="n">iloc</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
 
-  <span class="n">pooler</span> <span class="o">=</span> <span class="n">Pooler</span><span class="p">(</span>
+  <span class="n">pooler</span> <span class="o">=</span> <span class="n">CSVPooler</span><span class="p">(</span>
     <span class="s2">&quot;5mSalesRecords.csv&quot;</span><span class="p">,</span>
     <span class="n">process_dataframe</span><span class="p">,</span>
     <span class="n">callback_with</span><span class="o">=</span><span class="n">CallbackWith</span><span class="o">.</span><span class="n">CSV_FILENAME</span>
@@ -163,6 +163,10 @@ <h3 id="testing">Testing</h3>
 <pre><span></span><code><span class="w">  </span>clear<span class="p">;</span><span class="w"> </span>pytest
 </code></pre>
 </div>
+
+<h3 id="publishing">Publishing</h3>
+
+<p><code>poetry publish --build -u __token__ -p $PYPI_TOKEN</code></p>
 </div>
 
                         <input id="mod-csv_batcher-view-source" class="view-source-toggle-state" type="checkbox" aria-hidden="true" tabindex="-1">