From 0e450bd3f975bdeafefbb1c1e91425f4b5015d73 Mon Sep 17 00:00:00 2001
From: Abhishek Dasgupta <abhishek.dasgupta@dtc.ox.ac.uk>
Date: Thu, 18 Jan 2024 16:36:56 +0000
Subject: [PATCH] Add wordcloud_columns script

---
 src/wordcloud/requirements.txt     |  2 ++
 src/wordcloud/wordcloud_columns.py | 51 ++++++++++++++++++++++++++++++
 2 files changed, 53 insertions(+)
 create mode 100644 src/wordcloud/requirements.txt
 create mode 100644 src/wordcloud/wordcloud_columns.py

diff --git a/src/wordcloud/requirements.txt b/src/wordcloud/requirements.txt
new file mode 100644
index 0000000..739d645
--- /dev/null
+++ b/src/wordcloud/requirements.txt
@@ -0,0 +1,2 @@
+wordcloud==1.9.3
+pandas==2.*
diff --git a/src/wordcloud/wordcloud_columns.py b/src/wordcloud/wordcloud_columns.py
new file mode 100644
index 0000000..a811528
--- /dev/null
+++ b/src/wordcloud/wordcloud_columns.py
@@ -0,0 +1,51 @@
+"""
+Generate wordcloud from Pandas columns
+
+By default generates for all columns
+
+Use -e, --exclude with a comma separated list to
+exclude columns.
+
+Requirements: wordcloud pandas
+Install: pip install -r requirements.txt
+
+Usage:
+
+python wordcloud_columns.py <file> [--exclude=COLUMNS]
+"""
+
+import argparse
+from pathlib import Path
+
+import wordcloud
+import pandas as pd
+
+# Additional stopwords
+STOPWORDS = {"research", "University", "School", "Centre", "UK", "Institute"}
+
+# Set global wordcloud settings here
+# https://amueller.github.io/word_cloud/generated/wordcloud.WordCloud.html#wordcloud.WordCloud
+W = wordcloud.WordCloud(stopwords=set(wordcloud.STOPWORDS) | STOPWORDS)
+
+
+def generate_column(df, column: str):
+    assert column in df.columns, f"Column {column} not found in dataframe"
+    return W.generate(" ".join(df[column]))
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Generate wordcloud for data columns")
+    parser.add_argument("file", help="Filename to parse")
+    parser.add_argument("-e", "--exclude", help="Exclude columns")
+    args = parser.parse_args()
+
+    df = pd.read_csv(args.file)
+    columns = df.columns
+    excluded_columns = []
+    if args.exclude:
+        excluded_columns = args.exclude.split(",")
+    columns = [c for c in columns if c not in excluded_columns]
+    for c in columns:
+        output_filename = Path(args.file).stem + "_" + c.replace(" ", "-").replace(",", "") + ".jpg"
+        generate_column(df, c).to_file(output_filename)
+        print(output_filename)