zx-project/zx

A collection of experimental ZZ modules

zx

A collection of experimental ZZ modules.

Installation

In your project's zz.toml file:

[repos]
zx = "git://github.com/zx-project/zx.git"

Modules

zargv

Parse command line arguments supporting a variety of flag formats.

See the documentation for more information.

Usage

[dependencies]
zargv = "*"

using zargv

utils

Various utility functions, like memory management, etc.

See the documentation for more information.

Usage

[dependencies]
utils = "*"

using zxutils

curl

A ZZ binding to libcurl that makes working with curl in ZZ much easier.

See the documentation for more information.

Usage

[dependencies]
curl = "*"

using zxcurl

list

A doubly-linked list implementation.

See the documentation for more information.

Usage

[dependencies]
list = "*"

using list

http

A simple HTTP request library based on the [curl][curl] module.

See the documentation for more information.

Usage

[dependencies]
http = "*"

using zxhttp

See Also

• ZZ

License

MIT

zx's Issues

Documentation

ZZ Standard Library Upstream

Starting a discussion on modules created here or elsewhere that stand a chance to be upstreamed into the mainline ZZ standard module library

Use ZZ doc blocks

'child_process' module

References

• https://github.com/sheredom/subprocess.h
• http://docs.libuv.org/en/v1.x/guide/processes.html

'utf8' module

References:

• https://github.com/jwerle/libutf8
• https://github.com/sheredom/utf8.h

'tuv' binding

A libuv binding through Samsung's libuv fork/implementation for embedded systems if the project is maintained still and appears to work. Maybe not worth it though

See https://github.com/Samsung/libtuv

'libcsp' bindings

Reference: https://github.com/shiyanhui/libcsp

'zargv' review

https://github.com/zx-project/zx/tree/master/modules/zargv

cc @aep

some comments on list

unfortunately there's no inline review option unless you open a PR, so i'll comment this way

zx/modules/utils/src/math.zz

Line 1 in c3c2625

export inline fn max(u32 x, u32 y) -> u32 {

these should really be upstreamed

https://github.com/zx-project/zx/blob/master/modules/list/src/node.zz#L79

this exists

https://github.com/zetzit/zz/blob/master/modules/std/src/lib.zz#L10

but memory::zero sounds better than std::memset. should probably rename upstream

https://github.com/zx-project/zx/blob/master/modules/list/src/list.zz#L36

this probably meaningless. theories are very expensive (slow compilation) so don't use them excessively.

https://github.com/zx-project/zx/blob/master/modules/list/src/list.zz#L54

i like this struct, but for clarity head and tail should be marked unsafe, because they might be null.

https://github.com/zx-project/zx/blob/master/modules/list/src/list.zz#L115

(opinion) i think default() makes sense as function name here because you're filling in default values for the callbacks

https://github.com/zx-project/zx/blob/master/modules/list/src/list.zz#L106

does that mean this list can only store borrowed pointers? i mean, thats's actually a good idea.
i'm just scared of it because lifetimes aren't well defined in zetz yet. you could store a pointer to something that later gets deleted. anyway, that's an issue i need to work on, just be prepared for breaking changes in that direction :)

https://github.com/zx-project/zx/blob/master/modules/list/src/list.zz#L89

does that really have to be in a separate module? i'd rather have clean single file thingies with as little code as possible (more code = harder to read & more bugs)

https://github.com/zx-project/zx/blob/master/modules/list/src/iterator.zz#L58

dont return self pointers. that's confusing and currently makes both the prover explode as well as the nodejs export do very wrong things.

sorry that these things aren't useful compile errors. things are still moving around alot :)

https://github.com/zx-project/zx/blob/master/modules/list/src/iterator.zz#L140

i dont understand why this exists. why not just break?

https://github.com/zx-project/zx/blob/master/modules/list/src/iterator.zz#L174

destructors should be called drop(). like C++, drop() functions are called automatically when the lifetime ends (which is currently really broken, bear with me)

i'm not entirely sure what this does tho. generally structures that need no destructor are easier to manage

https://github.com/zx-project/zx/blob/master/modules/list/src/node.zz#L87

what does this do? deref usually refers to looking at a value refferenced by a pointer.

does the function remove a node from a list?

Rename `zxcurl` to `curlzz`

comments on curl

first of all this is super useful.
probably a reason for me to make sure we have a website with disoverable modules soon.

https://github.com/zx-project/zx/blob/master/modules/curl/zz.toml#L3

i mean, why not just "curl". you're the first one to make a curl module, so just grab the name ;)

https://github.com/zx-project/zx/blob/master/modules/curl/tests/curl.zz#L15

any chance to use the err module here? In general let me know how you think errors can be improved.

https://github.com/zx-project/zx/blob/master/modules/curl/src/easy.zz#L30

why are these nessesary? this looks like something that needs improving in zz?

https://github.com/zx-project/zx/blob/master/modules/curl/src/easy.zz#L68

this should require

where buffer_lenght <= len(buffer)

for safety.

https://github.com/zx-project/zx/blob/master/modules/curl/src/easy.zz#L110

should require where nullterm(string)

etc

https://github.com/zx-project/zx/blob/master/modules/curl/src/options.zz#L3

this looks tedious. why are these necessary?