Re^2: Modifying pos() from within (?{...})

Thanks for your help also!

But using unpack is better for, ahem, unpacking binary formats. You could write something like:

my ($tag, $data, $crc) = unpack q( (x4L X8Lx4/a* L)> ), $input;

Place ( )* around the template to get all the chunks.

Wow, that's concise and reasonably clear. I'm learning regex and more advanced pack|unpack at the same time so it's definitely something I need to dig into deeper.

Armed with the pack documentation I dissected your pack string into the following:

x4 - four null bytes (effectively skip the length fields)

L - 32bit value (the tag)

X8 - seek backwards 8 bytes (back to the start of the chunk

L - 32bit value (data length)

x4 - four null bytes (skip the tag field)

/ - pops the last value (data length) off the extraction stack and uses it as a repetition count for the next item

a* - binary data (count is set by the / above); not sure why * follows it, it works when removed

L - 32 bit CRC

It seems that I need to go through perlpacktut next!

I do have a small question however, when I try to repeat the group with * I get the following error:

'x' outside of string in unpack

Maybe this is something to do with the x at the start of the group? It's trying to read a new group starting just past the end of the string?

edit:Changing the first part of the pack string to L seemed to fix the 'outside of string' error. The data length is now explicitly within the returned values, but could be ignored later.

edit2:I span this out into a separate thread as it was getting too far off the original topic.

Comment on Re^2: Modifying pos() from within (?{...}) Select or Download Code

Replies are listed 'Best First'.
Re^3: Modifying pos() from within (?{...}) by Eily (Monsignor) on Apr 27, 2018 at 15:10 UTC
I already answered in the second thread, that the $input has an extra \n at the end, and I just realized why. The funny thing is, I reached that conclusion for the wrong reason. When I did my test I didn't modify $/ and didn't call chomp, so the \n wasn't removed. You, on the other hand did have chomp, but with $/ set to undef. The chomp doc states: When in slurp mode ($/ = undef ) or fixed-length record mode ($/ is a reference to an integer or the like; see perlvar), chomp won't remove anything. So in my tests by forgetting to call chomp, I got the same result you had by disabling its effect, and therefore reached the correct conclusion :D	[reply]
Re^4: Modifying pos() from within (?{...}) by AnomalousMonk (Archbishop) on Apr 27, 2018 at 16:31 UTC
When in slurp mode ... chomp won't remove anything. Another illustration of the wisdom of local-ized modification of globals in the narrowest scope, e.g. `my $input = do { local $/; <DATA>; };` rather than in broad or, indeed, global scope even when "it's just a small program; it doesn't really matter what we do to globals": spooky action at a distance. Give a man a fish: `<%-{-{-{-<`	[reply] [d/l] [select]