log_analyzer.py

from collections import Counter
import string

LOG_FILE = "logs/keystrokes.log"
WORD_FILE = "logs/word_analysis.log"
TIMING_FILE = "logs/timing_analysis.log"

"""
Generator that yields (header,lines) for each session
in order do respect the DRY principle and ensure cleaner code
"""
def get_sessions():
    current_header = None
    current_lines= []

    with open(LOG_FILE, "r", encoding="utf-8") as log_file:
        for line in log_file:
            if line.startswith("--- Session started"):
                if current_header:
                    yield current_header, current_lines
                current_header = line.strip()
                current_lines = []
            else:
                if line.strip():
                    current_lines.append(line)
        if current_header:
            yield current_header, current_lines

def analyze_words():
    with open(WORD_FILE, "w", encoding="utf-8") as w:
        for header, lines in get_sessions():
            counter = Counter()
            current_word = ""

            for line in lines:

                parts = line.split(maxsplit=1)
                if len(parts) < 2:
                    continue

                key = parts[1].strip()

                # we don't need [space] , [enter] etc. for the analysis
                if key.startswith("["):
                    cleaned = " "
                else:
                    cleaned = key

                if cleaned == " ":
                    if current_word:
                        counter.update([current_word])
                    current_word = ""
                else:
                    cleaned=cleaned.strip(string.punctuation) #don't want to count "word." separately from "word"
                    if cleaned:
                        current_word += cleaned
            if current_word:
                counter.update([current_word])
            # write results for this session
            w.write(header + "\n")
            for word, count in counter.most_common():
                w.write(f"{word},{count}\n")
            w.write("\n")


def analyze_timing():
    with open(TIMING_FILE, "w", encoding="utf-8") as ti:
        for header, lines in get_sessions():
            timestamps=[]
            for line in lines:
                parts=line.split(maxsplit=1)
                if len(parts) < 2:
                    continue
                t = float(parts[0])
                timestamps.append(t)

            total_keys = len(timestamps)
            ti.write(header + "\n")
            ti.write(f"Total keys: {total_keys}\n")

            if total_keys == 0:
                ti.write("No keystrokes recorded\n\n")
                continue

            session_duration = timestamps[-1] - timestamps[0]
            ti.write(f"Session duration: {session_duration:.3f}s\n")

            if total_keys == 1:
                ti.write("Not enough data for timing statistics\n\n")
                continue

            intervals = [
                timestamps[i] - timestamps[i - 1]
                for i in range(1, total_keys)
            ]

            avg_interval = sum(intervals) / len(intervals)
            fastest = min(intervals)
            slowest = max(intervals)

            keys_per_second = total_keys / session_duration if session_duration > 0 else 0
            pause_count = sum(1 for i in intervals if i > 1.0)

            ti.write(f"Avg interval: {avg_interval:.3f}s\n")
            ti.write(f"Fastest key: {fastest:.3f}s\n")
            ti.write(f"Slowest key: {slowest:.3f}s\n")
            ti.write(f"Typing speed: {keys_per_second:.2f} keys/s\n")
            ti.write(f"Pause count (>1s): {pause_count}\n\n")



def analyze():
    analyze_words()
    analyze_timing()