# from Adam Kennedy
# on Tuesday 05 January 2010 19:20:
>The one main issue I can think of regarding JSON is that we need to
>make sure we have a streaming parser, so that we don't need the entire
>thing in memory. We already have some memory issues with CPAN on just
>the highest-version index, and if we start pushing around all-version
>indexes this will just get worse.
Not to mention file size from repeating the field names on each entry
vs one header line and a non-negligible number of quote characters.
Applying a general-purpose parser with heavy quoting semantics to
simple tabular data probably loses the speed contest too.
BTW, sqlite has builtin tsv support.
--Eric
--
"Beware of bugs in the above code; I have only proved it correct, not
tried it."
--Donald Knuth
---------------------------------------------------
http://scratchcomputing.com
---------------------------------------------------
