clean up python scripts

Author: ritamtbsilva

.gitignore

@@ -18,3 +18,4 @@ markeR.Rproj
 inst/doc
 data_aux
 markeR.Rcheck 
+/python/.venv

python/README.md

@@ -12,14 +12,20 @@ Bioconductor R package **markeR** via `rpy2`.
 
 ## Prerequisites
 
-* R (>=.4.5) installed and on your `PATH`.
-* A Python virtual environment.  Install dependencies with:
-
-```bash
-python -m venv .venv
-source .venv/bin/activate
-pip install -r requirements.txt
-```
+* R (>=4.5) installed and on your `PATH`.
+* A Python virtual environment.  A `requirements.txt` file is provided in
+  this folder listing the needed packages (`rpy2`, `pandas`, `numpy` plus
+  optional `ipython`/`jupyter` for notebook usage).
+  To set up the environment:
+
+  ```bash
+  python -m venv .venv
+  source .venv/bin/activate
+  pip install -r requirements.txt
+  ```
+
+  After activation you can run the helper scripts using `python` from the
+  same environment.
 
 ## Quick start
 

python/markeR_to_python.py

@@ -60,12 +60,6 @@
 # converting.  helpers below wrap the recommended API.
 
 
-def _to_r(obj):
-    """Convert a pandas object to an R object using the current converter."""
-    with conversion.localconverter(ro.default_converter + pandas2ri.converter):
-        return conversion.py2rpy(obj)
-
-
 def _to_py(obj):
     """Convert an R object to a pandas/numpy equivalent."""
     with conversion.localconverter(ro.default_converter + pandas2ri.converter):
@@ -175,7 +169,6 @@ def plot_r_function(func_name: str, *args, width=800, height=600, filename=None,
 
 def ensure_bioc_installed() -> None:
     """Install Bioconductor's package manager if it is not already present."""
-    utils = importr("utils")
     biocinstaller = "BiocManager"
     if not isinstalled(biocinstaller):
         ro.r('install.packages("{0}")'.format(biocinstaller))
@@ -189,22 +182,17 @@ def install_markeR() -> None:
     the package should be loadable via `importr("markeR")`.
     """
     ensure_bioc_installed()
-    # use importr to check presence rather than only the isinstalled helper
-    try:
-        importr("markeR")
-    except Exception:
-        # attempt installation if import failed
+    if not isinstalled("markeR"):
         ro.r('BiocManager::install("markeR", ask=FALSE, update=FALSE)')
-        # load into namespace for side effects
-        ro.r('library(markeR)')
+    ro.r('library(markeR)')
 
 
 def get_markeR_functions() -> ro.Environment:
     """Return the markeR namespace so that functions can be accessed conveniently.
 
     Example:
         mark = get_markeR_functions()
-        imputed = mark.rgImpute(...)
+        scores = mark.CalculateScores(data=counts, metadata=metadata, gene_sets=genesets, method="logmedian")
     """
     install_markeR()
     # importing via importr is more reliable than accessing `ro.r['markeR']`.
@@ -330,4 +318,4 @@ def tutorial_benchmark(output_file=None):
             print("                            example: python markeR_to_python.py --tutorial --output my_plot.png")
     else:
         print("usage: python markeR_to_python.py --tutorial [--output FILENAME]")
-        print("See the module docstring for more details.")
+        print("See the module docstring for more details.")

python/requirements.txt

@@ -0,0 +1,9 @@
+# install packages for markeR_to_python script:
+#   pip install -r requirements.txt
+
+rpy2>=3.6
+pandas
+numpy
+# optional (for notebook inline display)
+ipython
+jupyter

python/run_marker_function.py

@@ -25,6 +25,7 @@
 import argparse
 import json
 import os
+import re
 
 # Check dependencies
 _missing = []
@@ -42,7 +43,6 @@
 
 def ensure_bioc_installed() -> None:
     """Install Bioconductor's package manager if it is not already present."""
-    utils = importr("utils")
     biocinstaller = "BiocManager"
     if not isinstalled(biocinstaller):
         ro.r('install.packages("{0}")'.format(biocinstaller))
@@ -52,11 +52,9 @@ def ensure_bioc_installed() -> None:
 def install_markeR() -> None:
     """Install the markeR package from Bioconductor if not already installed."""
     ensure_bioc_installed()
-    try:
-        importr("markeR")
-    except Exception:
+    if not isinstalled("markeR"):
         ro.r('BiocManager::install("markeR", ask=FALSE, update=FALSE)')
-        ro.r('library(markeR)')
+    ro.r('library(markeR)')
 
 
 def load_example_data():
@@ -113,7 +111,8 @@ def parse_parameter(value: str):
     except (json.JSONDecodeError, ValueError):
         pass
 
-    # Default: treat as string
+    # Default: treat as string, escaping any internal quotes
+    value = value.replace('"', '\\"')
     return f'"{value}"'
 
 
@@ -272,6 +271,12 @@ def main():
         return
 
     func_name = sys.argv[1]
+    
+    # Validate function name to prevent code injection
+    if not re.match(r'^[A-Za-z][A-Za-z0-9_.]*$', func_name):
+        sys.exit(f"Error: invalid function name '{func_name}'. "
+                 "Function names must start with a letter and contain only letters, digits, dots or underscores.")
+    
     output_file = None
     width = 800
     height = 600
@@ -358,4 +363,4 @@ def main():
 
 
 if __name__ == "__main__":
-    main()
+    main()