Storable's freeze/thaw vs store/retrieve

[Dlamini]

Hi Monks,

I'm confused as to what the difference is between using freeze/thaw vs using store/retrieve in the Storable module. Why might you use one over the other? Finally, are either of these useful for writing incrementally? I have a loop that will probably take a few days to complete and would like to be able to write chunks so that if I stop the loop partway, I can avoid re-doing the portion that is already complete. I'm sure a DB would work here as well, but for other reasons, I'd like to avoid creating one if possible.

Thanks in advance

[BrowserUk]

1. difference is between using freeze/thaw vs using store/retrieve in the Storable module.

   freeze/thaw write-to & read-from scalars. Ie, in memory. store/retrieve write-to & read-from files.

   Obviously, in order for the frozen data to persist beyond the program run, you would need to write it to persistant storage anyway.

2. are either of these useful for writing incrementally?

   That kind of depends on the nature of the data you are storing; and/or how you choose to use them.

   You could store your data to a file periodically, each time writing everything you have so far and overwriting the same file each time.

   Or, if the data you wish to save becomes complete in discrete chunks, then you could save each chunk to disk as it becomes complete. But you would need to write each chunk to a separate file as storable disk files are written and read as complete entities;

   It is possible to use freeze/thaw to write multiple storable chunks to a single file and then thaw them back, but it requires you to do the file handling yourself, including adding extra code to delimit the individual chunks.

More information about the nature and quantity of the data might yield better answers.

[Dlamini]

Thanks! That's exactly what I was looking for. I didn't understand the implication that writing to a scalar just meant writing it to memory that gets lost at the end of the program.