Add zlib:compress/1 #1570
Add zlib:compress/1 #1570
Conversation
https://www.erlang.org/doc/apps/erts/zlib.html#compress/1 Signed-off-by: Jakub Gonet <[email protected]>
jakub-gonet
force-pushed
the
jgonet/zlib-compress
branch
from
0a8ee68 to
55d0eaf
Compare
| @@ -2411,59 +2421,69 @@ static term nif_erlang_float_to_list(Context *ctx, int argc, term argv[]) | |||
| return make_list_from_ascii_buf((uint8_t *) float_buf, len, ctx); | |||
| } | |||
|
|
|||
| static term nif_erlang_list_to_binary_1(Context *ctx, int argc, term argv[]) | |||
| static term list_to_binary(term list, term *ret, Context *ctx) | |||
There was a problem hiding this comment.
If we want to keep this approach I really suggest renaming this function to something like: list_to_binary_maybe_gc so the caller will be aware about the fact that after a call to this function, any non trivial term will be invalid.
| if (result != OK_ATOM) { | ||
| RAISE_ERROR(result); | ||
| } | ||
| to_compress = to_compress_binary; |
There was a problem hiding this comment.
I don't really like this approach: we are basically creating a binary (with the same content of the iolist that has been given as argument) on the caller heap that will turn into garbage very soon.
Simple approach: factor out a function that allocates and populates a C buffer.
Cool approach: using iolist folding fun for compressing input piece by piece, without copying data to a temporary buffer.
| ok = test_zlib_compress_iolist(), | ||
| 0. | ||
|
|
||
| test_zlib_compress_binary() -> |
There was a problem hiding this comment.
Let's also test errors, such as: zlib:compress([900]) and zlib:compress([{}]).
https://www.erlang.org/doc/apps/erts/zlib.html#compress/1
These changes are made under both the "Apache 2.0" and the "GNU Lesser General
Public License 2.1 or later" license terms (dual license).
SPDX-License-Identifier: Apache-2.0 OR LGPL-2.1-or-later