\&iterator = File::KDBX::Iterator->new(\&iterator);
-Blesses an iterator to augment it with buffering plus some useful utility methods.
+Bless an iterator to augment it with buffering plus some useful utility methods.
=cut
$item = $iterator->();
$item = $iterator->next(\&query);
- $item = $iterator->next([\'simple expression', @fields]);
Get the next item or C<undef> if there are no more items. If a query is passed, get the next matching item,
discarding any unmatching items before the matching item. Example:
=method unget
+ # Replace buffer:
$iterator->unget(\@items);
- $iterator->unget(...);
# OR equivalently
$iterator->(\@items);
- $iterator->(...);
-Replace the buffer or unshift one or more items to the current buffer.
+ # Unshift onto buffer:
+ $iterator->unget(@items);
+ # OR equivalently
+ $iterator->(@items);
+
+Replace the buffer (first form) or unshift one or more items to the current buffer (second form).
See L</Buffer>.
$iterator->each($method_name, ...);
-Get or act on the rest of the items. There are three forms:
+Get or act on the rest of the items. This method has three forms:
=for :list
1. Without arguments, C<each> returns a list of the rest of the items.
2. Pass a coderef to be called once per item, in order. Arguments to the coderef are the item itself (also
- C<$_>), its index number and then any extra arguments that were passed to C<each> after the coderef.
+ available as C<$_>), its index number and then any extra arguments that were passed to C<each> after the
+ coderef.
3. Pass a string that is the name of a method to be called on each object, in order. Any extra arguments
passed to C<each> after the method name are passed through to each method call. This form requires each
item be an object that C<can> the given method.
=method where
\&iterator = $iterator->grep(\&query);
- \&iterator = $iterator->grep([\'simple expression', @fields]);
+ \&iterator = $iterator->grep(sub($item) { ... });
Get a new iterator draining from an existing iterator but providing only items that pass a test or are matched
-by a query.
+by a query. In its basic form this method is very much like perl's built-in grep function, except for
+iterators.
+
+There are many examples of the various forms of this method at L<File::KDBX/QUERY>.
=cut
\&iterator = $iterator->map(\&code);
-Get a new iterator draining from an existing iterator but providing modified items.
+Get a new iterator draining from an existing iterator but providing modified items. In its basic form this
+method is very much like perl's built-in map function, except for iterators.
=cut
* C<collate> - If true, use B<Unicode::Collate> (if available), otherwise use perl built-ins (default: true)
* Any B<Unicode::Collate> option is also supported.
-C<sort_by> and C<order_by> are aliases.
-
B<NOTE:> This method drains the iterator completely and places the sorted items onto the buffer. See
L</CAVEATS>.
=for :list
* C<ascending> - Order ascending if true, descending otherwise (default: true)
-C<nsort_by> and C<norder_by> are aliases.
-
B<NOTE:> This method drains the iterator completely and places the sorted items onto the buffer. See
L</CAVEATS>.
Get a new iterator draining from an existing iterator but providing only a limited number of items.
-C<limit> as an alias for L<Iterator::Simple/"$iterator->head($count)">.
+C<limit> is an alias for L<< Iterator::Simple/"$iterator->head($count)" >>.
=cut
A buffered iterator compatible with and expanding upon L<Iterator::Simple>, this provides an easy way to
navigate a L<File::KDBX> database. The documentation for B<Iterator::Simple> documents functions and methods
-supported but this iterator that are not documented here, so consider that additional reading.
+supported by this iterator that are not documented here, so consider that additional reading.
=head2 Buffer
This iterator is buffered, meaning it can drain from an iterator subroutine under the hood, storing items
-temporarily to be accessed later. This allows features like L</peek> and L</sort> which might be useful in the
-context of KDBX databases which are normally pretty small so draining an iterator isn't cost-prohibitive.
+temporarily to be accessed later. This allows features like L</peek> and L</order_by> which might be useful in
+the context of KDBX databases which are normally pretty small so draining an iterator completely isn't
+cost-prohibitive in terms of memory usage.
The way this works is that if you call an iterator without arguments, it acts like a normal iterator. If you
call it with arguments, however, the arguments are added to the buffer. When called without arguments, the
buffer is drained before the iterator function is. Using L</unget> is equivalent to calling the iterator with
-arguments, and as L</next> is equivalent to calling the iterator without arguments.
+arguments, and L</next> is equivalent to calling the iterator without arguments.
=head1 CAVEATS
its own ancestor -- but I'm noting it here as a potential issue if you use this iterator class for other
things (which you probably shouldn't do).
+KDBX databases are always fully-loaded into memory anyway, so there's not a significant memory cost to
+draining an iterator completely.
+
=cut
projects / chaz / p5-File-KDBX / blobdiff