Torque3D/Engine/lib/assimp/contrib/zip
Areloch 6ade6f08ce Updated Assimp
Added initial behavior for ImageAssets to hold a list of GFX resources of different texture profiles to avoid mem leaks with incorrect-typed usages
Added function to ImageAsset to get best-fit asset, allowing for fallbacks if the requested assetID is not found
Added function to ShapeAsset to get best-fit asset, allowing for fallbacks if the requested assetID is not found
Disabled fields for dynamic and static shadowmap refresh rates
Moved noShape model to core/rendering/shapes to place it in a more logical module position
Added an include to avoid undefined type compile error and removed unneeded semicolon from zone code
Added call to reload probe textures when a reloadTextures call is made
Adjusted default directional light shadowmap settings to not be as extreme
Added utility function to probe manager to allow any class to request a 'best fit' list of probes that would affect a given location, allowing other classes such as fog or particles to utilize IBL. Also updated probeManager's forward rendering to utilize same function to reduce code duplication.
Shifted shape loader code to utilize assimp for loader consistency and testing
Changed render bin used for SSAO postfx so it runs at the right time
Made Core_Rendering module scan for assets
Updated loose file references to a number of assets to follow proper formatting
Refactored asset import code to follow a more consistent object heirarchy structure on importing assets, allowing more reliable cross-referencing between inbound items
Updated asset import logic for materials/images so that they properly utilize ImageType. Images correctly save out the assigned image type, materials reference the images' type to know what map slot they should be used in. Importer logic also updated to better find-and-add associated images based on type.
Cleaned up a bunch of old, outdated code in the asset importer
Added initial handling for in-place importing of files without needing to process them through the UI.
Added ability to edit module script from RMB context menu if torsion path is set
Updated list field code for variable inspector to utilize correct ownerObject field
2020-03-19 09:47:38 -05:00
..
cmake Upgrade to Assimp 5.0 2019-11-10 14:40:50 +00:00
src Updated Assimp 2020-03-19 09:47:38 -05:00
test Updated Assimp 2020-03-19 09:47:38 -05:00
.gitignore Upgrade to Assimp 5.0 2019-11-10 14:40:50 +00:00
.travis.sh Upgrade to Assimp 5.0 2019-11-10 14:40:50 +00:00
.travis.yml Upgrade to Assimp 5.0 2019-11-10 14:40:50 +00:00
appveyor.yml Upgrade to Assimp 5.0 2019-11-10 14:40:50 +00:00
CMakeLists.txt Updated Assimp 2020-03-19 09:47:38 -05:00
README.md Updated Assimp 2020-03-19 09:47:38 -05:00
UNLICENSE WIP of assimp integration. 2019-02-08 16:25:43 -06:00
zip.png WIP of assimp integration. 2019-02-08 16:25:43 -06:00

A portable (OSX/Linux/Windows), simple zip library written in C

This is done by hacking awesome miniz library and layering functions on top of the miniz v1.15 API.

Build Version

The Idea

... Some day, I was looking for zip library written in C for my project, but I could not find anything simple enough and lightweight. Everything what I tried required 'crazy mental gymnastics' to integrate or had some limitations or was too heavy. I hate frameworks, factories and adding new dependencies. If I must to install all those dependencies and link new library, I'm getting almost sick. I wanted something powerfull and small enough, so I could add just a few files and compile them into my project. And finally I found miniz. Miniz is a lossless, high performance data compression library in a single source file. I only needed simple interface to append buffers or files to the current zip-entry. Thanks to this feature I'm able to merge many files/buffers and compress them on-the-fly.

It was the reason, why I decided to write zip module on top of the miniz. It required a little bit hacking and wrapping some functions, but I kept simplicity. So, you can grab these 3 files and compile them into your project. I hope that interface is also extremely simple, so you will not have any problems to understand it.

Examples

  • Create a new zip archive with default compression level.
struct zip_t *zip = zip_open("foo.zip", ZIP_DEFAULT_COMPRESSION_LEVEL, 'w');
{
    zip_entry_open(zip, "foo-1.txt");
    {
        const char *buf = "Some data here...\0";
        zip_entry_write(zip, buf, strlen(buf));
    }
    zip_entry_close(zip);

    zip_entry_open(zip, "foo-2.txt");
    {
        // merge 3 files into one entry and compress them on-the-fly.
        zip_entry_fwrite(zip, "foo-2.1.txt");
        zip_entry_fwrite(zip, "foo-2.2.txt");
        zip_entry_fwrite(zip, "foo-2.3.txt");
    }
    zip_entry_close(zip);
}
zip_close(zip);
  • Append to the existing zip archive.
struct zip_t *zip = zip_open("foo.zip", ZIP_DEFAULT_COMPRESSION_LEVEL, 'a');
{
    zip_entry_open(zip, "foo-3.txt");
    {
        const char *buf = "Append some data here...\0";
        zip_entry_write(zip, buf, strlen(buf));
    }
    zip_entry_close(zip);
}
zip_close(zip);
  • Extract a zip archive into a folder.
int on_extract_entry(const char *filename, void *arg) {
    static int i = 0;
    int n = *(int *)arg;
    printf("Extracted: %s (%d of %d)\n", filename, ++i, n);

    return 0;
}

int arg = 2;
zip_extract("foo.zip", "/tmp", on_extract_entry, &arg);
  • Extract a zip entry into memory.
void *buf = NULL;
size_t bufsize;

struct zip_t *zip = zip_open("foo.zip", 0, 'r');
{
    zip_entry_open(zip, "foo-1.txt");
    {
        zip_entry_read(zip, &buf, &bufsize);
    }
    zip_entry_close(zip);
}
zip_close(zip);

free(buf);
  • Extract a zip entry into memory (no internal allocation).
unsigned char *buf;
size_t bufsize;

struct zip_t *zip = zip_open("foo.zip", 0, 'r');
{
    zip_entry_open(zip, "foo-1.txt");
    {
        bufsize = zip_entry_size(zip);
        buf = calloc(sizeof(unsigned char), bufsize);

        zip_entry_noallocread(zip, (void *)buf, bufsize);
    }
    zip_entry_close(zip);
}
zip_close(zip);

free(buf);
  • Extract a zip entry into memory using callback.
struct buffer_t {
    char *data;
    size_t size;
};

static size_t on_extract(void *arg, unsigned long long offset, const void *data, size_t size) {
    struct buffer_t *buf = (struct buffer_t *)arg;
    buf->data = realloc(buf->data, buf->size + size + 1);
    assert(NULL != buf->data);

    memcpy(&(buf->data[buf->size]), data, size);
    buf->size += size;
    buf->data[buf->size] = 0;

    return size;
}

struct buffer_t buf = {0};
struct zip_t *zip = zip_open("foo.zip", 0, 'r');
{
    zip_entry_open(zip, "foo-1.txt");
    {
        zip_entry_extract(zip, on_extract, &buf);
    }
    zip_entry_close(zip);
}
zip_close(zip);

free(buf.data);
  • Extract a zip entry into a file.
struct zip_t *zip = zip_open("foo.zip", 0, 'r');
{
    zip_entry_open(zip, "foo-2.txt");
    {
        zip_entry_fread(zip, "foo-2.txt");
    }
    zip_entry_close(zip);
}
zip_close(zip);
  • List of all zip entries
struct zip_t *zip = zip_open("foo.zip", 0, 'r');
int i, n = zip_total_entries(zip);
for (i = 0; i < n; ++i) {
    zip_entry_openbyindex(zip, i);
    {
        const char *name = zip_entry_name(zip);
        int isdir = zip_entry_isdir(zip);
        unsigned long long size = zip_entry_size(zip);
        unsigned int crc32 = zip_entry_crc32(zip);
    }
    zip_entry_close(zip);
}
zip_close(zip);

Bindings

Compile zip library as a dynamic library.

$ mkdir build
$ cd build
$ cmake -DBUILD_SHARED_LIBS=true ..
$ make

Go (cgo)

package main

/*
#cgo CFLAGS: -I../src
#cgo LDFLAGS: -L. -lzip
#include <zip.h>
*/
import "C"
import "unsafe"

func main() {
	path := C.CString("/tmp/go.zip")
	zip := C.zip_open(path, 6, 'w')

	entryname := C.CString("test")
	C.zip_entry_open(zip, entryname)

	content := "test content"
	buf := unsafe.Pointer(C.CString(content))
	bufsize := C.size_t(len(content))
	C.zip_entry_write(zip, buf, bufsize)

	C.zip_entry_close(zip)

	C.zip_close(zip)
}

Rust (ffi)

extern crate libc;
use std::ffi::CString;

#[repr(C)]
pub struct Zip {
    _private: [u8; 0],
}

#[link(name = "zip")]
extern "C" {
    fn zip_open(path: *const libc::c_char, level: libc::c_int, mode: libc::c_char) -> *mut Zip;
    fn zip_close(zip: *mut Zip) -> libc::c_void;

    fn zip_entry_open(zip: *mut Zip, entryname: *const libc::c_char) -> libc::c_int;
    fn zip_entry_close(zip: *mut Zip) -> libc::c_int;
    fn zip_entry_write(
        zip: *mut Zip,
        buf: *const libc::c_void,
        bufsize: libc::size_t,
    ) -> libc::c_int;
}

fn main() {
    let path = CString::new("/tmp/test.zip").unwrap();
    let mode: libc::c_char = 'w' as libc::c_char;

    let entryname = CString::new("test.txt").unwrap();
    let content = "test content\0";

    unsafe {
        let zip: *mut Zip = zip_open(path.as_ptr(), 5, mode);
        {
            zip_entry_open(zip, entryname.as_ptr());
            {
                let buf = content.as_ptr() as *const libc::c_void;
                let bufsize = content.len() as libc::size_t;
                zip_entry_write(zip, buf, bufsize);
            }
            zip_entry_close(zip);
        }
        zip_close(zip);
    }
}

Ruby (ffi)

Install ffi gem.

$ gem install ffi

Bind in your module.

require 'ffi'

module Zip
  extend FFI::Library
  ffi_lib "./libzip.#{::FFI::Platform::LIBSUFFIX}"

  attach_function :zip_open, [:string, :int, :char], :pointer
  attach_function :zip_close, [:pointer], :void

  attach_function :zip_entry_open, [:pointer, :string], :int
  attach_function :zip_entry_close, [:pointer], :void
  attach_function :zip_entry_write, [:pointer, :string, :int], :int
end

ptr = Zip.zip_open("/tmp/ruby.zip", 6, "w".bytes()[0])

status = Zip.zip_entry_open(ptr, "test")

content = "test content"
status = Zip.zip_entry_write(ptr, content, content.size())

Zip.zip_entry_close(ptr)
Zip.zip_close(ptr)

Python (cffi)

Install cffi package

$ pip install cffi

Bind in your package.

import ctypes.util
from cffi import FFI

ffi = FFI()
ffi.cdef("""
    struct zip_t *zip_open(const char *zipname, int level, char mode);
    void zip_close(struct zip_t *zip);

    int zip_entry_open(struct zip_t *zip, const char *entryname);
    int zip_entry_close(struct zip_t *zip);
    int zip_entry_write(struct zip_t *zip, const void *buf, size_t bufsize);
""")

Zip = ffi.dlopen(ctypes.util.find_library("zip"))

ptr = Zip.zip_open("/tmp/python.zip", 6, 'w')

status = Zip.zip_entry_open(ptr, "test")

content = "test content"
status = Zip.zip_entry_write(ptr, content, len(content))

Zip.zip_entry_close(ptr)
Zip.zip_close(ptr)

Ring

The language comes with RingZip based on this library

load "ziplib.ring"

new Zip {
    setFileName("myfile.zip")
    open("w")
    newEntry() {
        open("test.c")
        writefile("test.c")
        close()
    }
    close()
}

Contribution Rules/Coding Standards

No need to throw away your coding style, just do your best to follow default clang-format style. Apply clang-format to the source files before commit:

for file in $(git ls-files | \grep -E '\.(c|h)$' | \grep -v -- '#')
do
    clang-format -i $file
done