Update arg kwargs parsing to support komma-separated lists also

Author: timvink

mkdocs_table_reader_plugin/safe_eval.py

@@ -55,34 +55,59 @@ def safe_eval(string):
  return literal_eval(string)
 
 
-def parse_argkwarg(string: str):
+def parse_argkwarg(input_str: str):
  """
  Parses a string to detect both args and kwargs.
  
  Adapted code from 
  https://stackoverflow.com/questions/9305387/string-of-kwargs-to-kwargs
 
  Args:
- string (str): string with positional and keyword arguments
+ input_str (str): string with positional and keyword arguments
  
  Returns:
  args[List], kwargs[Dict]
  """
-
- argkwargs = re.split(r"(?<!\=)(?:,{1} )(?!\=)", string)
+ # below generated by copilot, validated by unit tests
+ segments = []
+ current_segment = ''
+ in_quotes = False
+ quote_char = ''
+ bracket_count = 0
+ 
+ for char in input_str:
+ if char in "\"'" and not in_quotes:
+ in_quotes = True
+ quote_char = char
+ elif char == quote_char and in_quotes:
+ in_quotes = False
+ quote_char = ''
+ elif char == '[':
+ bracket_count += 1
+ elif char == ']':
+ bracket_count -= 1
+ elif char == ',' and not in_quotes and bracket_count == 0:
+ segments.append(current_segment.strip())
+ current_segment = ''
+ continue
+ 
+ current_segment += char
+ 
+ segments.append(current_segment.strip()) # Add the last segment
+ # end code generated by copilot, validated by unit tests
 
  args = []
  kwargs = []
 
- for i in argkwargs:
+ for i in segments:
  i = i.strip()
  if "=" in i:
  kwargs.append(i)
  else:
  if len(kwargs) != 0:
  raise AssertionError(
  "[table-reader-plugin] Make sure the python in your reader tag is correct: Positional arguments follow keyword arguments in '%s'"
- % string
+ % input_str
  )
  args.append(literal_eval(i))
 

setup.py

@@ -5,7 +5,7 @@
 
 setup(
  name="mkdocs-table-reader-plugin",
- version="2.1.0",
+ version="2.2.0",
  description="MkDocs plugin to directly insert tables from files into markdown.",
  long_description=long_description,
  long_description_content_type="text/markdown",

tests/fixtures/mixed_quotation_marks/docs/file.csv

@@ -0,0 +1,4 @@
+A,B,C,D,E
+10,23,45,78,99
+34,56,87,12,37
+93,67,34,23,11%

tests/fixtures/mixed_quotation_marks/docs/index.md

@@ -0,0 +1,7 @@
+# Test page
+
+This test is related to this issue: https://github.com/timvink/mkdocs-table-reader-plugin/issues/29
+
+## Table with mixed quote usage
+
+{{ read_csv("file.csv", usecols=['A', 'B']) }}

tests/fixtures/mixed_quotation_marks/mkdocs.yml

@@ -0,0 +1,7 @@
+site_name: test git_table_reader site
+use_directory_urls: true
+
+plugins:
+ - search
+ - table-reader:
+ data_path: "docs"

tests/test_build.py

@@ -346,6 +346,19 @@ def test_wrong_path(tmp_path):
  )
 
  result = build_docs_setup(tmp_proj)
- assert result.exit_code == 1, "'mkdocs build' command failed"
+ assert result.exit_code == 1, "'mkdocs build' command succeeded but should have failed"
  assert "[table-reader-plugin]: Cannot find table file" in result.output
  assert "non_existing_table.csv" in result.output
+
+
+def test_mixed_quotation_marks(tmp_path):
+ tmp_proj = setup_clean_mkdocs_folder(
+ "tests/fixtures/mixed_quotation_marks/mkdocs.yml", tmp_path
+ )
+ result = build_docs_setup(tmp_proj)
+ assert result.exit_code == 0, "'mkdocs build' command failed"
+
+ # Make sure the file.csv is inserted
+ page_with_tag = tmp_proj / "site/index.html"
+ contents = page_with_tag.read_text()
+ assert re.search(r"56", contents)

tests/test_kwargs.py

@@ -1,6 +1,7 @@
 
 import pandas as pd 
 from mkdocs_table_reader_plugin.utils import get_keywords, kwargs_in_func, kwargs_not_in_func
+from mkdocs_table_reader_plugin.safe_eval import parse_argkwarg
 
 
 def test_kwargs():
@@ -12,3 +13,10 @@ def test_kwargs():
  assert kwargs_in_func(keywords, pd.read_csv) == {'sep' : ';'}
  assert kwargs_not_in_func(keywords, pd.read_csv) == {'hi' : 'there'}
 
+
+def test_parse_argkwarg():
+
+ assert parse_argkwarg("sep=';'") == ([], {'sep': ';'})
+ assert parse_argkwarg('"file.csv", usecols=["A", "B"]') == (['file.csv'], {'usecols': ['A', 'B']})
+ assert parse_argkwarg('"file.csv", usecols=[\'A\',\'B\']') == (['file.csv'], {'usecols': ['A', 'B']})
+ assert parse_argkwarg("'assets/tables/table_with_carriage_return.csv', sep = ','") == (['assets/tables/table_with_carriage_return.csv'], {'sep': ','})