chatgpt prompts and codes

README.md

@@ -1,2 +1,19 @@
-# drought-monitor
-ChatGPT written code to create drought monitoring video
+## ChatGPT written code to create drought monitoring video
+
+Used publicly available ChatGPT https://chat.openai.com/chat and the following prompts to generate the code
+
+## prompt1
+
+```Write a python code to output a list of URLs in a file where each URL is in the form https://droughtmonitor.unl.edu/data/png/<date>/<date>_usdm.png where <date> is in the form YYYYMMDD ranges from 20000104 to 20230314 incremented by 7 days```
+
+
+The prompt is shown in the [screenshot prompt1](prompt1.png). The output is captured in the file [generate_urls.py](generate_urls.py)
+
+There are state or region specific drought map images available in the site https://droughtmonitor.unl.edu/maps/maparchive.aspx. For example, to get the drought maps of california use the URL pattern `https://droughtmonitor.unl.edu/data/png/<date>/<date>_ca_trd.png`
+
+## prompt2
+```Write a python code which takes a file named urls.txt containing list of image urls as input and generates a mp4 video with those images. Give 0.1 sec delay between downloading subsequent images ```
+
+The prompt is shown in the [screenshot prompt2](prompt2.png). The output is captured in the file [create_video.py](create_video.py)
+
+It missed importing numpy which I had to manually add. I also added the print statement to monitor the progress.

create_video.py

@@ -0,0 +1,44 @@
+import cv2
+import numpy as np
+import urllib.request
+
+# Define the URLs file name
+file_name = 'urls.txt'
+
+# Define the video frame size and FPS
+width, height = 640, 360
+fps = 30
+
+# Create a VideoWriter object to save the video
+fourcc = cv2.VideoWriter_fourcc(*'mp4v')
+out = cv2.VideoWriter('output.mp4', fourcc, fps, (width, height))
+
+# Open the URLs file and iterate over the lines
+with open(file_name) as f:
+    for line in f:
+        # Remove any leading/trailing whitespaces and line breaks
+        url = line.strip()
+        try:
+            print("Processing", url)
+            # Download the image from the URL
+            with urllib.request.urlopen(url) as url_response:
+                img_array = bytearray(url_response.read())
+
+            # Convert the byte array to a numpy array and reshape it to the desired frame size
+            img = np.asarray(img_array, dtype=np.uint8)
+            img = cv2.imdecode(img, cv2.IMREAD_UNCHANGED)
+            img = cv2.resize(img, (width, height))
+
+            # Add the frame to the video
+            out.write(img)
+
+            # Wait for 0.1 seconds before downloading the next image
+            cv2.waitKey(100)
+
+        except:
+            # If there is an error in downloading the image, skip it and continue with the next URL
+            continue
+
+# Release the VideoWriter object and destroy any remaining windows
+out.release()
+cv2.destroyAllWindows()

generate_urls.py

@@ -0,0 +1,24 @@
+import datetime
+
+# Define start and end dates
+start_date = datetime.date(2000, 1, 4)
+end_date = datetime.date(2023, 3, 14)
+
+# Define the URL format string
+url_template = 'https://droughtmonitor.unl.edu/data/png/{date}/{date}_usdm.png'
+
+# Define the output file name
+output_file = 'urls.txt'
+
+# Open the file for writing
+with open(output_file, 'w') as f:
+    # Loop over dates in 7-day increments
+    delta = datetime.timedelta(days=7)
+    d = start_date
+    while d <= end_date:
+        # Format the URL for the current date
+        url = url_template.format(date=d.strftime('%Y%m%d'))
+        # Write the URL to the file
+        f.write(url + '\n')
+        # Increment the date
+        d += delta