Line numbers appear to be offset

Python Help
1

I’ve gotten a fairly complicated error, but what I don’t understand is the lines its reporting as errors. The stack trace is checking if __name == "__main__", an empty line, opening a file, iterating over a list, assigning a variable, and sleeping (sleeping makes sense, but this is the wrong one. Logging indicates it was sleeping in a line 5 lines later)? It looks like the line numbers python is using are off, probably 5 lines earlier than they should be. By which I mean the line shown does correspond to the line number indicated, but I think that does not the line it should be displaying. What could be causing this? The code largely works fine, I just killed it with ^C.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/mnt/c/users/Max/Documents/python/Save_To_Wayback/savetowayback.py", line 585, in <module>
    if __name__ == "__main__":
     ^^^^^^
  File "/mnt/c/users/Max/Documents/python/Save_To_Wayback/savetowayback.py", line 574, in main

  File "/mnt/c/users/Max/Documents/python/Save_To_Wayback/savetowayback.py", line 556, in parse_args_and_save
    with open_utf8(NEW_URLS, "r") as file:
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/mnt/c/users/Max/Documents/python/Save_To_Wayback/savetowayback.py", line 508, in save_url_list
    for url in urls:

  File "/mnt/c/users/Max/Documents/python/Save_To_Wayback/savetowayback.py", line 485, in add_link
    last_url = None
     ^^^^^^^^^^^^^^
  File "/mnt/c/users/Max/Documents/python/Save_To_Wayback/savetowayback.py", line 476, in save_url
    time.sleep(BLOCKED_BY_ROBOTS_DELAY)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
KeyboardInterrupt

I’m not entirely sure what parts of my code to include, but here’s a couple potentially relevant parts.

The function where the error occurred:

def save_url(link: WebsiteLink) -> None:
	errors = 0
	while True:
		try:
			capture_with_logging(link)
			time.sleep(DEFAULT_DELAY)
			return
		except save.BlockedByRobots as exc:
			logger.error(f"Error {errors} Skipping blocked by robots: {link}, {exc}")
			# should not save in this case
			time.sleep(BLOCKED_BY_ROBOTS_DELAY)
			return
		except Exception as exc:
			errors += 1
			logger.warning(f"Error {errors}: {link}, {exc}")
			time.sleep(too_many_reqs_delay(errors))

The surrounding function of the second level:

def add_link(url: str) -> str | None:
	logger.debug(f"add_link {url}")
	last_url = None

	link = make_link(url.strip())

	while link.url:
		save_url(link)
		last_url = pick_url_to_save(link, url)
		link = link.attempt_get_next()

	return last_url
2

This can happen if you edit the code while it is running, forget to save before restarting etc. When Python shows the stack trace, it will look at the actual .py file in order to figure out the source code to show; and that can get out of sync with the compiled code that it actually executed.

This stack trace - ending in KeyboardInterrupt - is just what it looks like when you do that. It does not indicate any error in the code. (To elaborate: Python implements “killing it with ^C” by raising that exception; the idea is that a normal program won’t try to catch that, so it ends up killed by an uncaught exception.)

1 Like