useful-scripts/syncBlog.sh

#!/bin/bash
set -Eeuo pipefail

### USAGE ###
#./syncBlog.sh --obsidian-dir "/path/to/obsidian" --blog-dir "/path/to/blog"
###

# Get the directory of the script
SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"

# Default values
DEFAULT_OBSIDIAN_DIR="/Users/alex/Notes"
DEFAULT_BLOG_DIR="/Users/alex/Projects/alexhyett-blog"

# Parse command line arguments
while [[ $# -gt 0 ]]; do
    case $1 in
        --obsidian-dir)
            OBSIDIAN_DIR="$2"
            shift 2
            ;;
        --blog-dir)
            BLOG_DIR="$2"
            shift 2
            ;;
        *)
            echo "Unknown option: $1"
            exit 1
            ;;
    esac
done

# Set directory paths, using defaults if not provided
OBSIDIAN_DIR="${OBSIDIAN_DIR:-$DEFAULT_OBSIDIAN_DIR}"
BLOG_DIR="${BLOG_DIR:-$DEFAULT_BLOG_DIR}"

WEBSITE_DIR="$OBSIDIAN_DIR/6. WEBSITE/"
NOTES_DIR="$OBSIDIAN_DIR/5. GARDEN/"
LAST_RUN_FILE="$SCRIPT_DIR/.last_run"

# Print the directories being used
echo "Using Obsidian directory: $OBSIDIAN_DIR"
echo "Using Blog directory: $BLOG_DIR"
# Function to log messages with timestamps
log() {
    echo "$(date '+%Y-%m-%d %H:%M:%S') $1"
}

# Function to get the last run date
get_last_run_date() {
    if [[ -f "$LAST_RUN_FILE" ]]; then
        cat "$LAST_RUN_FILE"
    else
        echo "1970-01-01T00:00:00"
    fi
}

# Function to update the last run date
update_last_run_date() {
    date +"%Y-%m-%dT%H:%M:%S" > "$LAST_RUN_FILE"
}

# Function to check if a file has been updated since last run
is_file_updated() {
    local file=$1
    local last_run=$(get_last_run_date)
    local updated=$(sed -n 's/^updated: //p' "$file")
    
    if [[ -z "$updated" ]]; then
        return 0  # If no updated date, assume it needs processing
    fi
    
    if [[ "$updated" > "$last_run" ]]; then
        return 0  # File is newer
    else
        return 1  # File is older
    fi
}

# Function to copy updated files
copy_updated_files() {
    local source_dir=$1
    local dest_dir=$2
    local file_count=0

    while IFS= read -r -d '' file; do
        if is_file_updated "$file"; then
            mkdir -p "$(dirname "$dest_dir/${file#$source_dir}")"
            cp "$file" "$dest_dir/${file#$source_dir}"
            ((file_count++))
        fi
    done < <(find "$source_dir" -type f -name "*.md" -print0)

    echo "$file_count"
}

##### SCRIPT START #####
mkdir -p $BLOG_DIR/_staging

# Copy updated website files to staging
files=$(copy_updated_files "$WEBSITE_DIR" "$BLOG_DIR/_staging")
if [[ "$files" != "0" ]]; then
    log "Transferred $files updated website files to staging"
fi

# Copy updated notes files to staging
files=$(copy_updated_files "$NOTES_DIR" "$BLOG_DIR/_staging/notes")
if [[ "$files" != "0" ]]; then
    log "Transferred $files updated notes files to staging"
fi

# Function to extract the permalink from YAML frontmatter
extract_permalink() {
    local note_file=$1
    grep "^permalink:" "$note_file" | sed 's/^permalink: //'
}

# Function to process each note file
process_notes() {
    local note_file=$1
    local debug_file=${note_file/$BLOG_DIR\/_staging/""}
    local reference_notes_dir=$2
    local temp_file=$(mktemp)
    local modified=false

    # Read through each line in the note file
    while IFS= read -r line || [[ -n "$line" ]]; do
        # Use a simpler regex pattern to find wiki links in the line
        link_regex='\[\[([^|[]+)(\|([^[]+))?\]\]'
        
        # Replace links in the line
        new_line=""
        rest_of_line="$line"

        while [[ $rest_of_line =~ $link_regex ]]; do
            full_match="${BASH_REMATCH[0]}"
            file_name="${BASH_REMATCH[1]}"
            display_name="${BASH_REMATCH[3]:-$file_name}"
            before="${rest_of_line%%"$full_match"*}"
            after="${rest_of_line#*"$full_match"}"
            
            # Find the reference note file
            note_full_path=$(find "$reference_notes_dir" -type f -iname "$file_name.md" | head -n 1)

            echo $note_full_path
            
            replacement="$full_match" # Default replacement is the original link

            # Extract the permalink if the reference note file exists
            if [[ -n $note_full_path ]]; then
                permalink=$(extract_permalink "$note_full_path")
                if [[ -n $permalink ]]; then
                    log "$debug_file: Found permalink: $permalink"
                    replacement="[$display_name]($permalink)"
                    modified=true
                else
                    log "$debug_file: No permalink found for: $note_full_path"
                fi
            else
               log "$debug_file: No matching file found for: $file_name"
            fi

            new_line+="$before$replacement"
            rest_of_line="$after"
        done
        
        new_line+="$rest_of_line"
        
        # Write the processed line to the temp_file
        echo "$new_line" >> "$temp_file"
    done < "$note_file"

    # If modifications were made, overwrite the original file; otherwise, delete the temp_file
    if $modified; then
        mv "$temp_file" "$note_file"
    else
        rm "$temp_file"
    fi
}

log "Starting note processing"

# Recursively find and process each Markdown note in the directory
find "$BLOG_DIR/_staging" -type f -name "*.md" | while IFS= read -r note_file; do
    process_notes "$note_file" "$BLOG_DIR/content"
done

log "Finished note processing"

# Sync converted files to the blog directory
output=$(rsync -zarvh --stats "$BLOG_DIR/_staging/" "$BLOG_DIR/content/" | grep -A 1 'files transferred')
files=$(echo "$output" | awk '/files transferred/{print $NF}')

if [[ "$files" != "0" ]]; then
    log "Transferred $files files to blog"
fi

# Delete _staging dir
rm -rf $BLOG_DIR/_staging

# Update the last run date
update_last_run_date

log "Script completed successfully"