OXIESEC PANEL

Name	Size	Modified	Perms
📁 ..	-	01/29/2025 10:07:18 PM	rwxr-xr-x
📄 FlexLexer.h	6.73 KB	01/30/2022 08:23:38 AM	rw-r--r--
📄 GeoIP.h	17.42 KB	01/17/2018 07:23:18 PM	rw-r--r--
📄 GeoIPCity.h	2.24 KB	01/17/2018 07:23:18 PM	rw-r--r--
📄 a.out.h	4.25 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 aio.h	7.56 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 aliases.h	1.98 KB	03/10/2025 10:35:11 AM	rw-r--r--
📄 alloca.h	1.17 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 ar.h	1.69 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 argp.h	24.95 KB	03/10/2025 10:35:10 AM	rw-r--r--
📄 argz.h	5.91 KB	03/10/2025 10:34:47 AM	rw-r--r--
📁 arpa	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 asm	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📁 asm-generic	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📄 assert.h	4.46 KB	03/10/2025 10:34:44 AM	rw-r--r--
📁 bits	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 bsock	-	08/27/2024 05:43:51 PM	rwxr-xr-x
📄 byteswap.h	1.42 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 complex.h	7.95 KB	03/10/2025 10:34:45 AM	rw-r--r--
📄 cpio.h	2.21 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 cpuidle.h	931 bytes	06/11/2025 12:06:00 PM	rw-r--r--
📁 criu	-	08/27/2024 05:56:55 PM	rwxr-xr-x
📄 crypt.h	10.9 KB	02/10/2022 04:05:00 AM	rw-r--r--
📄 ctype.h	10.71 KB	03/10/2025 10:34:44 AM	rw-r--r--
📄 dirent.h	12.22 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 dlfcn.h	7.52 KB	03/10/2025 10:34:46 AM	rw-r--r--
📁 drm	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📄 elf.h	178.14 KB	03/10/2025 10:35:12 AM	rw-r--r--
📄 endian.h	2.25 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 envz.h	2.8 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 err.h	2.29 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 errno.h	1.64 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 error.h	2.36 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 execinfo.h	1.49 KB	03/10/2025 10:35:10 AM	rw-r--r--
📄 fcntl.h	11.17 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 features-time64.h	1.37 KB	03/10/2025 10:34:42 AM	rw-r--r--
📄 features.h	17.69 KB	03/10/2025 10:34:42 AM	rw-r--r--
📄 fenv.h	5.65 KB	03/10/2025 10:34:45 AM	rw-r--r--
📁 finclude	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📄 fmtmsg.h	3.16 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 fnmatch.h	2.24 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 fpu_control.h	3.5 KB	03/10/2025 10:34:45 AM	rw-r--r--
📄 fstab.h	3.04 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 fts.h	9.35 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 ftw.h	6.19 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 gconv.h	4.11 KB	03/10/2025 10:34:43 AM	rw-r--r--
📄 gelf.h	11.14 KB	03/01/2024 08:12:17 PM	rw-r--r--
📄 getopt.h	1.43 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 glob.h	7.13 KB	03/10/2025 10:34:48 AM	rw-r--r--
📁 gnu	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📄 gnu-versions.h	2.29 KB	03/10/2025 10:34:42 AM	rw-r--r--
📄 gnumake.h	2.84 KB	01/03/2020 07:11:27 AM	rw-r--r--
📄 grp.h	6.53 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 gshadow.h	4.42 KB	03/10/2025 10:35:10 AM	rw-r--r--
📄 iconv.h	1.81 KB	03/10/2025 10:34:43 AM	rw-r--r--
📄 ieee754.h	4.8 KB	03/10/2025 10:34:45 AM	rw-r--r--
📄 ifaddrs.h	2.77 KB	03/10/2025 10:35:11 AM	rw-r--r--
📄 inttypes.h	8.14 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 langinfo.h	17.43 KB	03/10/2025 10:34:43 AM	rw-r--r--
📄 lastlog.h	126 bytes	03/10/2025 10:35:12 AM	rw-r--r--
📄 lauxlib.h	9.1 KB	09/26/2023 06:28:33 PM	rw-r--r--
📄 libelf.h	19.84 KB	03/01/2024 08:12:17 PM	rw-r--r--
📄 libgen.h	1.35 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 libintl.h	4.47 KB	03/10/2025 10:34:44 AM	rw-r--r--
📄 liblsapi-sha1.h	569 bytes	02/21/2025 10:42:38 AM	rw-r--r--
📁 libxml2	-	03/12/2025 06:46:00 PM	rwxr-xr-x
📄 limits.h	5.57 KB	03/10/2025 10:34:42 AM	rw-r--r--
📄 link.h	7.05 KB	03/10/2025 10:35:12 AM	rw-r--r--
📁 linux	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📄 locale.h	7.5 KB	03/10/2025 10:34:43 AM	rw-r--r--
📄 lsapidef.h	4.6 KB	02/21/2025 10:42:38 AM	rw-r--r--
📄 lscapi.h	24.21 KB	02/21/2025 10:42:38 AM	rw-r--r--
📄 lscapi_config.h	599 bytes	02/21/2025 10:42:38 AM	rw-r--r--
📄 lua.h	15.45 KB	09/26/2023 06:28:33 PM	rw-r--r--
📄 lua.hpp	191 bytes	09/26/2023 06:28:33 PM	rw-r--r--
📄 luaconf-x86_64.h	21.01 KB	09/26/2023 06:28:33 PM	rw-r--r--
📄 luaconf.h	1.62 KB	09/26/2023 05:46:00 PM	rw-r--r--
📄 lualib.h	1.09 KB	09/26/2023 06:28:33 PM	rw-r--r--
📁 lzma	-	08/27/2024 05:53:19 PM	rwxr-xr-x
📄 lzma.h	9.63 KB	03/17/2020 02:28:50 PM	rw-r--r--
📄 malloc.h	5.77 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 math.h	47.63 KB	03/10/2025 10:34:45 AM	rw-r--r--
📄 mcheck.h	2.38 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 memory.h	956 bytes	03/10/2025 10:34:47 AM	rw-r--r--
📁 misc	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📄 mntent.h	3.28 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 monetary.h	1.92 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 mqueue.h	4.5 KB	03/10/2025 10:34:47 AM	rw-r--r--
📁 mtd	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📁 mysql	-	08/27/2024 05:35:53 PM	rwxr-xr-x
📁 net	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 netash	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 netatalk	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 netax25	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📄 netdb.h	27.79 KB	03/10/2025 10:35:11 AM	rw-r--r--
📁 neteconet	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 netinet	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 netipx	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 netiucv	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 netpacket	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 netrom	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 netrose	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📁 nfs	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📄 nl_types.h	1.71 KB	03/10/2025 10:34:44 AM	rw-r--r--
📄 nlist.h	1.56 KB	03/01/2024 08:12:17 PM	rw-r--r--
📄 nss.h	14.07 KB	03/10/2025 10:35:11 AM	rw-r--r--
📄 obstack.h	20.81 KB	03/10/2025 10:34:47 AM	rw-r--r--
📁 openssl	-	04/07/2025 07:53:04 PM	rwxr-xr-x
📄 paths.h	2.91 KB	03/10/2025 10:34:48 AM	rw-r--r--
📁 pcp	-	01/29/2025 10:11:54 PM	rwxr-xr-x
📄 poll.h	22 bytes	03/10/2025 10:34:48 AM	rw-r--r--
📄 powercap.h	1.62 KB	06/11/2025 12:06:00 PM	rw-r--r--
📄 printf.h	6.71 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 proc_service.h	3.4 KB	03/10/2025 10:35:11 AM	rw-r--r--
📁 protocols	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📄 pthread.h	47.24 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 pty.h	1.53 KB	03/10/2025 10:35:12 AM	rw-r--r--
📄 pwd.h	6.17 KB	03/10/2025 10:34:48 AM	rw-r--r--
📁 python3.9	-	04/17/2025 10:53:24 AM	rwxr-xr-x
📁 rdma	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📄 re_comp.h	963 bytes	03/10/2025 10:34:48 AM	rw-r--r--
📄 regex.h	25.3 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 regexp.h	1.41 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 resolv.h	12.02 KB	03/10/2025 10:35:11 AM	rw-r--r--
📁 rpc	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📄 sched.h	4.92 KB	03/10/2025 10:34:48 AM	rw-r--r--
📁 scsi	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📄 search.h	5.32 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 semaphore.h	3.38 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 setjmp.h	3.12 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 sgtty.h	1.31 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 shadow.h	5.34 KB	03/10/2025 10:35:10 AM	rw-r--r--
📄 signal.h	12.73 KB	03/10/2025 10:34:46 AM	rw-r--r--
📁 sound	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📄 spawn.h	7.84 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 stab.h	264 bytes	03/10/2025 10:34:48 AM	rw-r--r--
📄 stdc-predef.h	2.24 KB	03/10/2025 10:34:42 AM	rw-r--r--
📄 stdint.h	8.28 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 stdio.h	30.67 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 stdio_ext.h	2.73 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 stdlib.h	35.46 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 string.h	18.33 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 strings.h	4.64 KB	03/10/2025 10:34:47 AM	rw-r--r--
📁 sys	-	03/18/2025 09:44:14 PM	rwxr-xr-x
📄 syscall.h	25 bytes	03/10/2025 10:34:48 AM	rw-r--r--
📄 sysexits.h	5.11 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 syslog.h	24 bytes	03/10/2025 10:34:48 AM	rw-r--r--
📄 tar.h	3.7 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 termio.h	214 bytes	03/10/2025 10:34:48 AM	rw-r--r--
📄 termios.h	3.51 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 tgmath.h	39.24 KB	03/10/2025 10:34:45 AM	rw-r--r--
📄 thread_db.h	15.65 KB	03/10/2025 10:35:11 AM	rw-r--r--
📄 threads.h	7.51 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 time.h	14.49 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 ttyent.h	2.44 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 uchar.h	1.96 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 ucontext.h	1.99 KB	03/10/2025 10:34:46 AM	rw-r--r--
📄 ulimit.h	1.55 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 unistd.h	43.45 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 utime.h	1.86 KB	03/10/2025 10:34:48 AM	rw-r--r--
📄 utmp.h	3.15 KB	03/10/2025 10:35:12 AM	rw-r--r--
📄 utmpx.h	4 KB	03/10/2025 10:35:12 AM	rw-r--r--
📄 values.h	1.91 KB	03/10/2025 10:34:42 AM	rw-r--r--
📁 video	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📄 wait.h	22 bytes	03/10/2025 10:34:48 AM	rw-r--r--
📄 wchar.h	31.39 KB	03/10/2025 10:34:47 AM	rw-r--r--
📄 wctype.h	5.42 KB	03/10/2025 10:34:49 AM	rw-r--r--
📄 wordexp.h	2.44 KB	03/10/2025 10:34:48 AM	rw-r--r--
📁 xen	-	06/20/2025 01:51:24 AM	rwxr-xr-x
📄 zconf.h	15.88 KB	09/26/2023 09:22:15 AM	rw-r--r--
📄 zdict.h	25.03 KB	12/20/2021 10:49:18 PM	rw-r--r--
📄 zlib.h	94 KB	09/26/2023 09:22:15 AM	rw-r--r--
📄 zstd.h	145.16 KB	12/20/2021 10:49:18 PM	rw-r--r--
📄 zstd_errors.h	3.73 KB	12/20/2021 10:49:18 PM	rw-r--r--

Editing: zdict.h

/*
 * Copyright (c) Yann Collet, Facebook, Inc.
 * All rights reserved.
 *
 * This source code is licensed under both the BSD-style license (found in the
 * LICENSE file in the root directory of this source tree) and the GPLv2 (found
 * in the COPYING file in the root directory of this source tree).
 * You may select, at your option, one of the above-listed licenses.
 */

#ifndef DICTBUILDER_H_001
#define DICTBUILDER_H_001

#if defined (__cplusplus)
extern "C" {
#endif

/*======  Dependencies  ======*/
#include <stddef.h>  /* size_t */

/* =====   ZDICTLIB_API : control library symbols visibility   ===== */
#ifndef ZDICTLIB_VISIBILITY
#  if defined(__GNUC__) && (__GNUC__ >= 4)
#    define ZDICTLIB_VISIBILITY __attribute__ ((visibility ("default")))
#  else
#    define ZDICTLIB_VISIBILITY
#  endif
#endif
#if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1)
#  define ZDICTLIB_API __declspec(dllexport) ZDICTLIB_VISIBILITY
#elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1)
#  define ZDICTLIB_API __declspec(dllimport) ZDICTLIB_VISIBILITY /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/
#else
#  define ZDICTLIB_API ZDICTLIB_VISIBILITY
#endif

/*******************************************************************************
 * Zstd dictionary builder
 *
 * FAQ
 * ===
 * Why should I use a dictionary?
 * ------------------------------
 *
 * Zstd can use dictionaries to improve compression ratio of small data.
 * Traditionally small files don't compress well because there is very little
 * repetition in a single sample, since it is small. But, if you are compressing
 * many similar files, like a bunch of JSON records that share the same
 * structure, you can train a dictionary on ahead of time on some samples of
 * these files. Then, zstd can use the dictionary to find repetitions that are
 * present across samples. This can vastly improve compression ratio.
 *
 * When is a dictionary useful?
 * ----------------------------
 *
 * Dictionaries are useful when compressing many small files that are similar.
 * The larger a file is, the less benefit a dictionary will have. Generally,
 * we don't expect dictionary compression to be effective past 100KB. And the
 * smaller a file is, the more we would expect the dictionary to help.
 *
 * How do I use a dictionary?
 * --------------------------
 *
 * Simply pass the dictionary to the zstd compressor with
 * `ZSTD_CCtx_loadDictionary()`. The same dictionary must then be passed to
 * the decompressor, using `ZSTD_DCtx_loadDictionary()`. There are other
 * more advanced functions that allow selecting some options, see zstd.h for
 * complete documentation.
 *
 * What is a zstd dictionary?
 * --------------------------
 *
 * A zstd dictionary has two pieces: Its header, and its content. The header
 * contains a magic number, the dictionary ID, and entropy tables. These
 * entropy tables allow zstd to save on header costs in the compressed file,
 * which really matters for small data. The content is just bytes, which are
 * repeated content that is common across many samples.
 *
 * What is a raw content dictionary?
 * ---------------------------------
 *
 * A raw content dictionary is just bytes. It doesn't have a zstd dictionary
 * header, a dictionary ID, or entropy tables. Any buffer is a valid raw
 * content dictionary.
 *
 * How do I train a dictionary?
 * ----------------------------
 *
 * Gather samples from your use case. These samples should be similar to each
 * other. If you have several use cases, you could try to train one dictionary
 * per use case.
 *
 * Pass those samples to `ZDICT_trainFromBuffer()` and that will train your
 * dictionary. There are a few advanced versions of this function, but this
 * is a great starting point. If you want to further tune your dictionary
 * you could try `ZDICT_optimizeTrainFromBuffer_cover()`. If that is too slow
 * you can try `ZDICT_optimizeTrainFromBuffer_fastCover()`.
 *
 * If the dictionary training function fails, that is likely because you
 * either passed too few samples, or a dictionary would not be effective
 * for your data. Look at the messages that the dictionary trainer printed,
 * if it doesn't say too few samples, then a dictionary would not be effective.
 *
 * How large should my dictionary be?
 * ----------------------------------
 *
 * A reasonable dictionary size, the `dictBufferCapacity`, is about 100KB.
 * The zstd CLI defaults to a 110KB dictionary. You likely don't need a
 * dictionary larger than that. But, most use cases can get away with a
 * smaller dictionary. The advanced dictionary builders can automatically
 * shrink the dictionary for you, and select a the smallest size that
 * doesn't hurt compression ratio too much. See the `shrinkDict` parameter.
 * A smaller dictionary can save memory, and potentially speed up
 * compression.
 *
 * How many samples should I provide to the dictionary builder?
 * ------------------------------------------------------------
 *
 * We generally recommend passing ~100x the size of the dictionary
 * in samples. A few thousand should suffice. Having too few samples
 * can hurt the dictionaries effectiveness. Having more samples will
 * only improve the dictionaries effectiveness. But having too many
 * samples can slow down the dictionary builder.
 *
 * How do I determine if a dictionary will be effective?
 * -----------------------------------------------------
 *
 * Simply train a dictionary and try it out. You can use zstd's built in
 * benchmarking tool to test the dictionary effectiveness.
 *
 *   # Benchmark levels 1-3 without a dictionary
 *   zstd -b1e3 -r /path/to/my/files
 *   # Benchmark levels 1-3 with a dictionary
 *   zstd -b1e3 -r /path/to/my/files -D /path/to/my/dictionary
 *
 * When should I retrain a dictionary?
 * -----------------------------------
 *
 * You should retrain a dictionary when its effectiveness drops. Dictionary
 * effectiveness drops as the data you are compressing changes. Generally, we do
 * expect dictionaries to "decay" over time, as your data changes, but the rate
 * at which they decay depends on your use case. Internally, we regularly
 * retrain dictionaries, and if the new dictionary performs significantly
 * better than the old dictionary, we will ship the new dictionary.
 *
 * I have a raw content dictionary, how do I turn it into a zstd dictionary?
 * -------------------------------------------------------------------------
 *
 * If you have a raw content dictionary, e.g. by manually constructing it, or
 * using a third-party dictionary builder, you can turn it into a zstd
 * dictionary by using `ZDICT_finalizeDictionary()`. You'll also have to
 * provide some samples of the data. It will add the zstd header to the
 * raw content, which contains a dictionary ID and entropy tables, which
 * will improve compression ratio, and allow zstd to write the dictionary ID
 * into the frame, if you so choose.
 *
 * Do I have to use zstd's dictionary builder?
 * -------------------------------------------
 *
 * No! You can construct dictionary content however you please, it is just
 * bytes. It will always be valid as a raw content dictionary. If you want
 * a zstd dictionary, which can improve compression ratio, use
 * `ZDICT_finalizeDictionary()`.
 *
 * What is the attack surface of a zstd dictionary?
 * ------------------------------------------------
 *
 * Zstd is heavily fuzz tested, including loading fuzzed dictionaries, so
 * zstd should never crash, or access out-of-bounds memory no matter what
 * the dictionary is. However, if an attacker can control the dictionary
 * during decompression, they can cause zstd to generate arbitrary bytes,
 * just like if they controlled the compressed data.
 *
 ******************************************************************************/

/*! ZDICT_trainFromBuffer():
 *  Train a dictionary from an array of samples.
 *  Redirect towards ZDICT_optimizeTrainFromBuffer_fastCover() single-threaded, with d=8, steps=4,
 *  f=20, and accel=1.
 *  Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
 *  supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
 *  The resulting dictionary will be saved into `dictBuffer`.
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *  Note:  Dictionary training will fail if there are not enough samples to construct a
 *         dictionary, or if most of the samples are too small (< 8 bytes being the lower limit).
 *         If dictionary training fails, you should use zstd without a dictionary, as the dictionary
 *         would've been ineffective anyways. If you believe your samples would benefit from a dictionary
 *         please open an issue with details, and we can look into it.
 *  Note: ZDICT_trainFromBuffer()'s memory usage is about 6 MB.
 *  Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
 *        It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
 *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
 *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
 */
ZDICTLIB_API size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity,
                                    const void* samplesBuffer,
                                    const size_t* samplesSizes, unsigned nbSamples);

typedef struct {
    int      compressionLevel;   /*< optimize for a specific zstd compression level; 0 means default */
    unsigned notificationLevel;  /*< Write log to stderr; 0 = none (default); 1 = errors; 2 = progression; 3 = details; 4 = debug; */
    unsigned dictID;             /*< force dictID value; 0 means auto mode (32-bits random value)
                                  *   NOTE: The zstd format reserves some dictionary IDs for future use.
                                  *         You may use them in private settings, but be warned that they
                                  *         may be used by zstd in a public dictionary registry in the future.
                                  *         These dictionary IDs are:
                                  *           - low range  : <= 32767
                                  *           - high range : >= (2^31)
                                  */
} ZDICT_params_t;

/*! ZDICT_finalizeDictionary():
 * Given a custom content as a basis for dictionary, and a set of samples,
 * finalize dictionary by adding headers and statistics according to the zstd
 * dictionary format.
 *
 * Samples must be stored concatenated in a flat buffer `samplesBuffer`,
 * supplied with an array of sizes `samplesSizes`, providing the size of each
 * sample in order. The samples are used to construct the statistics, so they
 * should be representative of what you will compress with this dictionary.
 *
 * The compression level can be set in `parameters`. You should pass the
 * compression level you expect to use in production. The statistics for each
 * compression level differ, so tuning the dictionary for the compression level
 * can help quite a bit.
 *
 * You can set an explicit dictionary ID in `parameters`, or allow us to pick
 * a random dictionary ID for you, but we can't guarantee no collisions.
 *
 * The dstDictBuffer and the dictContent may overlap, and the content will be
 * appended to the end of the header. If the header + the content doesn't fit in
 * maxDictSize the beginning of the content is truncated to make room, since it
 * is presumed that the most profitable content is at the end of the dictionary,
 * since that is the cheapest to reference.
 *
 * `maxDictSize` must be >= max(dictContentSize, ZSTD_DICTSIZE_MIN).
 *
 * @return: size of dictionary stored into `dstDictBuffer` (<= `maxDictSize`),
 *          or an error code, which can be tested by ZDICT_isError().
 * Note: ZDICT_finalizeDictionary() will push notifications into stderr if
 *       instructed to, using notificationLevel>0.
 * NOTE: This function currently may fail in several edge cases including:
 *         * Not enough samples
 *         * Samples are uncompressible
 *         * Samples are all exactly the same
 */
ZDICTLIB_API size_t ZDICT_finalizeDictionary(void* dstDictBuffer, size_t maxDictSize,
                                const void* dictContent, size_t dictContentSize,
                                const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
                                ZDICT_params_t parameters);

/*======   Helper functions   ======*/
ZDICTLIB_API unsigned ZDICT_getDictID(const void* dictBuffer, size_t dictSize);  /**< extracts dictID; @return zero if error (not a valid dictionary) */
ZDICTLIB_API size_t ZDICT_getDictHeaderSize(const void* dictBuffer, size_t dictSize);  /* returns dict header size; returns a ZSTD error code on failure */
ZDICTLIB_API unsigned ZDICT_isError(size_t errorCode);
ZDICTLIB_API const char* ZDICT_getErrorName(size_t errorCode);

#ifdef ZDICT_STATIC_LINKING_ONLY

/* ====================================================================================
 * The definitions in this section are considered experimental.
 * They should never be used with a dynamic library, as they may change in the future.
 * They are provided for advanced usages.
 * Use them only in association with static linking.
 * ==================================================================================== */

#define ZDICT_DICTSIZE_MIN    256
/* Deprecated: Remove in v1.6.0 */
#define ZDICT_CONTENTSIZE_MIN 128

/*! ZDICT_cover_params_t:
 *  k and d are the only required parameters.
 *  For others, value 0 means default.
 */
typedef struct {
    unsigned k;                  /* Segment size : constraint: 0 < k : Reasonable range [16, 2048+] */
    unsigned d;                  /* dmer size : constraint: 0 < d <= k : Reasonable range [6, 16] */
    unsigned steps;              /* Number of steps : Only used for optimization : 0 means default (40) : Higher means more parameters checked */
    unsigned nbThreads;          /* Number of threads : constraint: 0 < nbThreads : 1 means single-threaded : Only used for optimization : Ignored if ZSTD_MULTITHREAD is not defined */
    double splitPoint;           /* Percentage of samples used for training: Only used for optimization : the first nbSamples * splitPoint samples will be used to training, the last nbSamples * (1 - splitPoint) samples will be used for testing, 0 means default (1.0), 1.0 when all samples are used for both training and testing */
    unsigned shrinkDict;         /* Train dictionaries to shrink in size starting from the minimum size and selects the smallest dictionary that is shrinkDictMaxRegression% worse than the largest dictionary. 0 means no shrinking and 1 means shrinking  */
    unsigned shrinkDictMaxRegression; /* Sets shrinkDictMaxRegression so that a smaller dictionary can be at worse shrinkDictMaxRegression% worse than the max dict size dictionary. */
    ZDICT_params_t zParams;
} ZDICT_cover_params_t;

typedef struct {
    unsigned k;                  /* Segment size : constraint: 0 < k : Reasonable range [16, 2048+] */
    unsigned d;                  /* dmer size : constraint: 0 < d <= k : Reasonable range [6, 16] */
    unsigned f;                  /* log of size of frequency array : constraint: 0 < f <= 31 : 1 means default(20)*/
    unsigned steps;              /* Number of steps : Only used for optimization : 0 means default (40) : Higher means more parameters checked */
    unsigned nbThreads;          /* Number of threads : constraint: 0 < nbThreads : 1 means single-threaded : Only used for optimization : Ignored if ZSTD_MULTITHREAD is not defined */
    double splitPoint;           /* Percentage of samples used for training: Only used for optimization : the first nbSamples * splitPoint samples will be used to training, the last nbSamples * (1 - splitPoint) samples will be used for testing, 0 means default (0.75), 1.0 when all samples are used for both training and testing */
    unsigned accel;              /* Acceleration level: constraint: 0 < accel <= 10, higher means faster and less accurate, 0 means default(1) */
    unsigned shrinkDict;         /* Train dictionaries to shrink in size starting from the minimum size and selects the smallest dictionary that is shrinkDictMaxRegression% worse than the largest dictionary. 0 means no shrinking and 1 means shrinking  */
    unsigned shrinkDictMaxRegression; /* Sets shrinkDictMaxRegression so that a smaller dictionary can be at worse shrinkDictMaxRegression% worse than the max dict size dictionary. */

ZDICT_params_t zParams;
} ZDICT_fastCover_params_t;

/*! ZDICT_trainFromBuffer_cover():
 *  Train a dictionary from an array of samples using the COVER algorithm.
 *  Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
 *  supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
 *  The resulting dictionary will be saved into `dictBuffer`.
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 *  Note: ZDICT_trainFromBuffer_cover() requires about 9 bytes of memory for each input byte.
 *  Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
 *        It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
 *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
 *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
 */
ZDICTLIB_API size_t ZDICT_trainFromBuffer_cover(
          void *dictBuffer, size_t dictBufferCapacity,
    const void *samplesBuffer, const size_t *samplesSizes, unsigned nbSamples,
          ZDICT_cover_params_t parameters);

/*! ZDICT_optimizeTrainFromBuffer_cover():
 * The same requirements as above hold for all the parameters except `parameters`.
 * This function tries many parameter combinations and picks the best parameters.
 * `*parameters` is filled with the best parameters found,
 * dictionary constructed with those parameters is stored in `dictBuffer`.
 *
 * All of the parameters d, k, steps are optional.
 * If d is non-zero then we don't check multiple values of d, otherwise we check d = {6, 8}.
 * if steps is zero it defaults to its default value.
 * If k is non-zero then we don't check multiple values of k, otherwise we check steps values in [50, 2000].
 *
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          On success `*parameters` contains the parameters selected.
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 * Note: ZDICT_optimizeTrainFromBuffer_cover() requires about 8 bytes of memory for each input byte and additionally another 5 bytes of memory for each byte of memory for each thread.
 */
ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_cover(
          void* dictBuffer, size_t dictBufferCapacity,
    const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
          ZDICT_cover_params_t* parameters);

/*! ZDICT_trainFromBuffer_fastCover():
 *  Train a dictionary from an array of samples using a modified version of COVER algorithm.
 *  Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
 *  supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
 *  d and k are required.
 *  All other parameters are optional, will use default values if not provided
 *  The resulting dictionary will be saved into `dictBuffer`.
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 *  Note: ZDICT_trainFromBuffer_fastCover() requires 6 * 2^f bytes of memory.
 *  Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
 *        It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
 *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
 *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
 */
ZDICTLIB_API size_t ZDICT_trainFromBuffer_fastCover(void *dictBuffer,
                    size_t dictBufferCapacity, const void *samplesBuffer,
                    const size_t *samplesSizes, unsigned nbSamples,
                    ZDICT_fastCover_params_t parameters);

/*! ZDICT_optimizeTrainFromBuffer_fastCover():
 * The same requirements as above hold for all the parameters except `parameters`.
 * This function tries many parameter combinations (specifically, k and d combinations)
 * and picks the best parameters. `*parameters` is filled with the best parameters found,
 * dictionary constructed with those parameters is stored in `dictBuffer`.
 * All of the parameters d, k, steps, f, and accel are optional.
 * If d is non-zero then we don't check multiple values of d, otherwise we check d = {6, 8}.
 * if steps is zero it defaults to its default value.
 * If k is non-zero then we don't check multiple values of k, otherwise we check steps values in [50, 2000].
 * If f is zero, default value of 20 is used.
 * If accel is zero, default value of 1 is used.
 *
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          On success `*parameters` contains the parameters selected.
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 * Note: ZDICT_optimizeTrainFromBuffer_fastCover() requires about 6 * 2^f bytes of memory for each thread.
 */
ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_fastCover(void* dictBuffer,
                    size_t dictBufferCapacity, const void* samplesBuffer,
                    const size_t* samplesSizes, unsigned nbSamples,
                    ZDICT_fastCover_params_t* parameters);

typedef struct {
    unsigned selectivityLevel;   /* 0 means default; larger => select more => larger dictionary */
    ZDICT_params_t zParams;
} ZDICT_legacy_params_t;

/*! ZDICT_trainFromBuffer_legacy():
 *  Train a dictionary from an array of samples.
 *  Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
 *  supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
 *  The resulting dictionary will be saved into `dictBuffer`.
 * `parameters` is optional and can be provided with values set to 0 to mean "default".
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 *  Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
 *        It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
 *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
 *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
 *  Note: ZDICT_trainFromBuffer_legacy() will send notifications into stderr if instructed to, using notificationLevel>0.
 */
ZDICTLIB_API size_t ZDICT_trainFromBuffer_legacy(
    void* dictBuffer, size_t dictBufferCapacity,
    const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
    ZDICT_legacy_params_t parameters);

/* Deprecation warnings */
/* It is generally possible to disable deprecation warnings from compiler,
   for example with -Wno-deprecated-declarations for gcc
   or _CRT_SECURE_NO_WARNINGS in Visual.
   Otherwise, it's also possible to manually define ZDICT_DISABLE_DEPRECATE_WARNINGS */
#ifdef ZDICT_DISABLE_DEPRECATE_WARNINGS
#  define ZDICT_DEPRECATED(message) ZDICTLIB_API   /* disable deprecation warnings */
#else
#  define ZDICT_GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
#  if defined (__cplusplus) && (__cplusplus >= 201402) /* C++14 or greater */
#    define ZDICT_DEPRECATED(message) [[deprecated(message)]] ZDICTLIB_API
#  elif defined(__clang__) || (ZDICT_GCC_VERSION >= 405)
#    define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated(message)))
#  elif (ZDICT_GCC_VERSION >= 301)
#    define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated))
#  elif defined(_MSC_VER)
#    define ZDICT_DEPRECATED(message) ZDICTLIB_API __declspec(deprecated(message))
#  else
#    pragma message("WARNING: You need to implement ZDICT_DEPRECATED for this compiler")
#    define ZDICT_DEPRECATED(message) ZDICTLIB_API
#  endif
#endif /* ZDICT_DISABLE_DEPRECATE_WARNINGS */

ZDICT_DEPRECATED("use ZDICT_finalizeDictionary() instead")
size_t ZDICT_addEntropyTablesFromBuffer(void* dictBuffer, size_t dictContentSize, size_t dictBufferCapacity,
                                  const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples);

#endif   /* ZDICT_STATIC_LINKING_ONLY */

#if defined (__cplusplus)
}
#endif

#endif   /* DICTBUILDER_H_001 */