writing

mostly bash/gitbash-compliant ports of Python scripts to have some automations on restricted Windows machines

find bold text (**) in markdown files

find . -type f -name "*.md" | while read -r file; do
    grep -oE '\*\*[^*]+\*\*' "$file" \
    | sed 's/^\*\*//; s/\*\*$//' \
    | sort -u
done # > ./bold-phrases.txt

find Title Case phrases in markdown files

...with awk!

Awk is around my age. And it's awesome. Give it some love (or at least give it some use).

Even wikipedia gives it a quick tutorial. Also check up on awk.info (last time I did, it's been for sale with some pretty ugly AI-generated images).

find . -type f -name "*.md" | while read -r file; do
    grep -P '^[^#><*].*\b[A-Z][a-zA-Z]*\b' "$file" \
    | awk '{
        for (i=2; i<=NF; i++) {
            if ($i ~ /^[A-Z][a-zA-Z]*$/) {
                entry = $i;
                while ((i+1)<=NF && $(i+1) ~ /^[A-Z][a-zA-Z]*$/) {
                    entry = entry " " $(i+1);
                    i++;
                }
                print entry;
            }
        }
    }' \
    | sort -u
done

populate/replace Alt text for images in markdown files

requires a csv:

filename,alt_text
main-config.png,Screenshot of the main configuration screen
...

Basically a wrapper for sed (with -E to ensure proper handling of escapes in pattern)

Click to see code...

#!/bin/bash

# Check args
if [ "$#" -ne 2 ]; then
    echo "Usage: $0 <csv_file> <md_files_directory>"
    exit 1
fi

csv_file="$1"
md_dir="$2"

# Check if CSV file exists
if [ ! -f "$csv_file" ]; then
    echo "Error: '$csv_file' not found."
    exit 1
fi

# Check if directory exists
if [ ! -d "$md_dir" ]; then
    echo "Error: '$md_dir' not found."
    exit 1
fi

# Read CSV
while IFS=',' read -r image_name alt_text; do
    # Skip empty lines/fields
    if [[ -z "$image_name" || -z "$alt_text" ]]; then
        continue
    fi
    alt_text=$(echo "$alt_text" | tr -d '\r\n')
    echo Adding: $alt_text
    echo to: $image_name
    echo ""

    # Escape special chars in image_name (for grep/sed)
    image_name_escaped=$(echo "$image_name" | sed -e 's/[\\&/]/\\\\&/g')

    # Find-replace in all .md files in the directory 
    find "$md_dir" -type f -name "*.md" -exec sed -E -i \
        "s/!\\[[^]]*\\]\\(([^)]*\/$image_name_escaped[^)]*)\\)/![${alt_text}](\1)/g" {} +

done < <(tail -n +2 "$csv_file") # Skip the header