{ "content": [ { "type": "text", "text": "# unzstd(1) (man)\n\n**Summary:** zstd - zstd, zstdmt, unzstd, zstdcat - Compress or decompress .zst files\n\n**Synopsis:** zstd [OPTIONS] [-|INPUT-FILE] [-o OUTPUT-FILE]\nzstdmt is equivalent to zstd -T0\nunzstd is equivalent to zstd -d\nzstdcat is equivalent to zstd -dcf\n\n## Flags\n\n| Flag | Long | Arg | Description |\n|------|------|-----|-------------|\n| -z | --compress | — | Compress. This is the default operation mode when no operation mode option is speci‐ fied and no other operation mode is |\n| -d | --uncompress | — | Decompress. |\n| -t | --test | — | Test the integrity of compressed files. This option is equivalent to --decompress --stdout except that the decompressed |\n| — | — | — | |\n| — | --train | — | Use FILEs as a training set to create a dictionary. The training set should contain a lot of small files (> 100). |\n| -l | --list | — | Display information related to a zstd compressed file, such as size, ratio, and check‐ sum. Some of these fields may not |\n\n## Section Outline\n\n- **NAME** (2 lines)\n- **SYNOPSIS** (8 lines)\n- **DESCRIPTION** (39 lines) — 1 subsections\n - Concatenation with .zst files (3 lines)\n- **OPTIONS** (1 lines) — 10 subsections\n - Integer suffixes and special values (8 lines)\n - Operation mode (2 lines)\n - -z --compress (4 lines)\n - -d --decompress --uncompress (2 lines)\n - -t --test (4 lines)\n - -b# (1 lines)\n - --train FILEs (3 lines)\n - -l --list (4 lines)\n - Operation modifiers (156 lines)\n - Restricted usage of Environment Variables (18 lines)\n- **DICTIONARY BUILDER** (6 lines) — 8 subsections\n - --train FILEs (9 lines)\n - -o file (2 lines)\n - --maxdict=# (6 lines)\n - -B# (1 lines)\n - --dictID=# (7 lines)\n - --train-cover[=k#,d=#,steps=#,split=#,shrink[=#]] (35 lines)\n - --train-fastcover[=k#,d=#,f=#,steps=#,split=#,accel=#] (17 lines)\n - --train-legacy[=selectivity=#] (11 lines)\n- **BENCHMARK** (1 lines) — 5 subsections\n - -b# (1 lines)\n - -e# -b# -e# (1 lines)\n - -i# (1 lines)\n - -B# --block-size=# (2 lines)\n - --priority=rt (10 lines)\n- **ADVANCED COMPRESSION OPTIONS** (1 lines) — 3 subsections\n - --zstd[=options]: (122 lines)\n - Example (5 lines)\n - -B#: (6 lines)\n- **BUGS** (2 lines)\n- **AUTHOR** (5 lines)\n\n## Full Content\n\n### NAME\n\nzstd - zstd, zstdmt, unzstd, zstdcat - Compress or decompress .zst files\n\n### SYNOPSIS\n\nzstd [OPTIONS] [-|INPUT-FILE] [-o OUTPUT-FILE]\n\nzstdmt is equivalent to zstd -T0\n\nunzstd is equivalent to zstd -d\n\nzstdcat is equivalent to zstd -dcf\n\n### DESCRIPTION\n\nzstd is a fast lossless compression algorithm and data compression tool, with command line\nsyntax similar to gzip (1) and xz (1). It is based on the LZ77 family, with further FSE &\nhuff0 entropy stages. zstd offers highly configurable compression speed, with fast modes at >\n200 MB/s per core, and strong modes nearing lzma compression ratios. It also features a very\nfast decoder, with speeds > 500 MB/s per core.\n\nzstd command line syntax is generally similar to gzip, but features the following differences\n:\n\n• Source files are preserved by default. It´s possible to remove them automatically by us‐\ning the --rm command.\n\n• When compressing a single file, zstd displays progress notifications and result summary\nby default. Use -q to turn them off.\n\n• zstd does not accept input from console, but it properly accepts stdin when it´s not the\nconsole.\n\n• zstd displays a short help page when command line is an error. Use -q to turn it off.\n\n\n\nzstd compresses or decompresses each file according to the selected operation mode. If no\nfiles are given or file is -, zstd reads from standard input and writes the processed data to\nstandard output. zstd will refuse to write compressed data to standard output if it is a ter‐\nminal : it will display an error message and skip the file. Similarly, zstd will refuse to\nread compressed data from standard input if it is a terminal.\n\nUnless --stdout or -o is specified, files are written to a new file whose name is derived\nfrom the source file name:\n\n• When compressing, the suffix .zst is appended to the source filename to get the target\nfilename.\n\n• When decompressing, the .zst suffix is removed from the source filename to get the target\nfilename\n\n#### Concatenation with .zst files\n\nIt is possible to concatenate .zst files as is. zstd will decompress such files as if they\nwere a single .zst file.\n\n### OPTIONS\n\n#### Integer suffixes and special values\n\nIn most places where an integer argument is expected, an optional suffix is supported to eas‐\nily indicate large integers. There must be no space between the integer and the suffix.\n\nKiB Multiply the integer by 1,024 (2^10). Ki, K, and KB are accepted as synonyms for KiB.\n\nMiB Multiply the integer by 1,048,576 (2^20). Mi, M, and MB are accepted as synonyms for\nMiB.\n\n#### Operation mode\n\nIf multiple operation mode options are given, the last one takes effect.\n\n#### -z --compress\n\nCompress. This is the default operation mode when no operation mode option is speci‐\nfied and no other operation mode is implied from the command name (for example, unzstd\nimplies --decompress).\n\n#### -d --decompress --uncompress\n\nDecompress.\n\n#### -t --test\n\nTest the integrity of compressed files. This option is equivalent to --decompress\n--stdout except that the decompressed data is discarded instead of being written to\nstandard output. No files are created or removed.\n\n#### -b#\n\n#### --train FILEs\n\nUse FILEs as a training set to create a dictionary. The training set should contain a\nlot of small files (> 100).\n\n#### -l --list\n\nDisplay information related to a zstd compressed file, such as size, ratio, and check‐\nsum. Some of these fields may not be available. This command can be augmented with the\n-v modifier.\n\n#### Operation modifiers\n\n• -#: # compression level [1-19] (default: 3)\n\n• --ultra: unlocks high compression levels 20+ (maximum 22), using a lot more memory. Note\nthat decompression will also require more memory when using these levels.\n\n• --fast[=#]: switch to ultra-fast compression levels. If =# is not present, it defaults to\n1. The higher the value, the faster the compression speed, at the cost of some compres‐\nsion ratio. This setting overwrites compression level if one was set previously. Simi‐\nlarly, if a compression level is set after --fast, it overrides it.\n\n• -T#, --threads=#: Compress using # working threads (default: 1). If # is 0, attempt to\ndetect and use the number of physical CPU cores. In all cases, the nb of threads is\ncapped to ZSTDMTNBWORKERSMAX==200. This modifier does nothing if zstd is compiled with‐\nout multithread support.\n\n• --single-thread: Does not spawn a thread for compression, use a single thread for both\nI/O and compression. In this mode, compression is serialized with I/O, which is slightly\nslower. (This is different from -T1, which spawns 1 compression thread in parallel of\nI/O). This mode is the only one available when multithread support is disabled. Sin‐\ngle-thread mode features lower memory usage. Final compressed result is slightly differ‐\nent from -T1.\n\n• --adapt[=min=#,max=#] : zstd will dynamically adapt compression level to perceived I/O\nconditions. Compression level adaptation can be observed live by using command -v. Adap‐\ntation can be constrained between supplied min and max levels. The feature works when\ncombined with multi-threading and --long mode. It does not work with --single-thread. It\nsets window size to 8 MB by default (can be changed manually, see wlog). Due to the\nchaotic nature of dynamic adaptation, compressed result is not reproducible. note : at\nthe time of this writing, --adapt can remain stuck at low speed when combined with multi‐\nple worker threads (>=2).\n\n• --long[=#]: enables long distance matching with # windowLog, if not # is not present it\ndefaults to 27. This increases the window size (windowLog) and memory usage for both the\ncompressor and decompressor. This setting is designed to improve the compression ratio\nfor files with long matches at a large distance.\n\nNote: If windowLog is set to larger than 27, --long=windowLog or --memory=windowSize\nneeds to be passed to the decompressor.\n\n• -D DICT: use DICT as Dictionary to compress or decompress FILE(s)\n\n• --patch-from FILE: Specify the file to be used as a reference point for zstd´s diff en‐\ngine. This is effectively dictionary compression with some convenient parameter selec‐\ntion, namely that windowSize > srcSize.\n\nNote: cannot use both this and -D together Note: --long mode will be automatically acti‐\nvated if chainLog < fileLog (fileLog being the windowLog required to cover the whole\nfile). You can also manually force it. Node: for all levels, you can use --patch-from in\n--single-thread mode to improve compression ratio at the cost of speed Note: for level\n19, you can get increased compression ratio at the cost of speed by specifying\n--zstd=targetLength= to be something large (i.e 4096), and by setting a large\n--zstd=chainLog=\n\n• --rsyncable : zstd will periodically synchronize the compression state to make the com‐\npressed file more rsync-friendly. There is a negligible impact to compression ratio, and\nthe faster compression levels will see a small compression speed hit. This feature does\nnot work with --single-thread. You probably don´t want to use it with long range mode,\nsince it will decrease the effectiveness of the synchronization points, but your milage\nmay vary.\n\n• -C, --[no-]check: add integrity check computed from uncompressed data (default: enabled)\n\n• --[no-]content-size: enable / disable whether or not the original size of the file is\nplaced in the header of the compressed file. The default option is --content-size (mean‐\ning that the original size will be placed in the header).\n\n• --no-dictID: do not store dictionary ID within frame header (dictionary compression). The\ndecoder will have to rely on implicit knowledge about which dictionary to use, it won´t\nbe able to check if it´s correct.\n\n• -M#, --memory=#: Set a memory usage limit. By default, Zstandard uses 128 MB for decom‐\npression as the maximum amount of memory the decompressor is allowed to use, but you can\noverride this manually if need be in either direction (ie. you can increase or decrease\nit).\n\nThis is also used during compression when using with --patch-from=. In this case, this\nparameter overrides that maximum size allowed for a dictionary. (128 MB).\n\n• --stream-size=# : Sets the pledged source size of input coming from a stream. This value\nmust be exact, as it will be included in the produced frame header. Incorrect stream\nsizes will cause an error. This information will be used to better optimize compression\nparameters, resulting in better and potentially faster compression, especially for\nsmaller source sizes.\n\n• --size-hint=#: When handling input from a stream, zstd must guess how large the source\nsize will be when optimizing compression parameters. If the stream size is relatively\nsmall, this guess may be a poor one, resulting in a higher compression ratio than ex‐\npected. This feature allows for controlling the guess when needed. Exact guesses result\nin better compression ratios. Overestimates result in slightly degraded compression ra‐\ntios, while underestimates may result in significant degradation.\n\n• -o FILE: save result into FILE\n\n• -f, --force: overwrite output without prompting, and (de)compress symbolic links\n\n• -c, --stdout: force write to standard output, even if it is the console\n\n• --[no-]sparse: enable / disable sparse FS support, to make files with many zeroes smaller\non disk. Creating sparse files may save disk space and speed up decompression by reducing\nthe amount of disk I/O. default: enabled when output is into a file, and disabled when\noutput is stdout. This setting overrides default and can force sparse mode over stdout.\n\n• --rm: remove source file(s) after successful compression or decompression. If used in\ncombination with -o, will trigger a confirmation prompt (which can be silenced with -f),\nas this is a destructive operation.\n\n• -k, --keep: keep source file(s) after successful compression or decompression. This is\nthe default behavior.\n\n• -r: operate recursively on directories\n\n• --filelist FILE read a list of files to process as content from FILE. Format is compati‐\nble with ls output, with one file per line.\n\n• --output-dir-flat DIR: resulting files are stored into target DIR directory, instead of\nsame directory as origin file. Be aware that this command can introduce name collision\nissues, if multiple files, from different directories, end up having the same name. Col‐\nlision resolution ensures first file with a given name will be present in DIR, while in\ncombination with -f, the last file will be present instead.\n\n• --output-dir-mirror DIR: similar to --output-dir-flat, the output files are stored under‐\nneath target DIR directory, but this option will replicate input directory hierarchy into\noutput DIR.\n\nIf input directory contains \"..\", the files in this directory will be ignored. If input\ndirectory is an absolute directory (i.e. \"/var/tmp/abc\"), it will be stored into the\n\"output-dir/var/tmp/abc\". If there are multiple input files or directories, name colli‐\nsion resolution will follow the same rules as --output-dir-flat.\n\n• --format=FORMAT: compress and decompress in other formats. If compiled with support, zstd\ncan compress to or decompress from other compression algorithm formats. Possibly avail‐\nable options are zstd, gzip, xz, lzma, and lz4. If no such format is provided, zstd is\nthe default.\n\n• -h/-H, --help: display help/long help and exit\n\n• -V, --version: display version number and exit. Advanced : -vV also displays supported\nformats. -vvV also displays POSIX support. -q will only display the version number, suit‐\nable for machine reading.\n\n• -v, --verbose: verbose mode, display more information\n\n• -q, --quiet: suppress warnings, interactivity, and notifications. specify twice to sup‐\npress errors too.\n\n• --no-progress: do not display the progress bar, but keep all other messages.\n\n• --show-default-cparams: Shows the default compression parameters that will be used for a\nparticular src file. If the provided src file is not a regular file (eg. named pipe), the\ncli will just output the default parameters. That is, the parameters that are used when\nthe src size is unknown.\n\n• --: All arguments after -- are treated as files\n\n#### Restricted usage of Environment Variables\n\nUsing environment variables to set parameters has security implications. Therefore, this av‐\nenue is intentionally restricted. Only ZSTDCLEVEL and ZSTDNBTHREADS are currently sup‐\nported. They set the compression level and number of threads to use during compression, re‐\nspectively.\n\nZSTDCLEVEL can be used to set the level between 1 and 19 (the \"normal\" range). If the value\nof ZSTDCLEVEL is not a valid integer, it will be ignored with a warning message. ZSTDCLEVEL\njust replaces the default compression level (3).\n\nZSTDNBTHREADS can be used to set the number of threads zstd will attempt to use during com‐\npression. If the value of ZSTDNBTHREADS is not a valid unsigned integer, it will be ignored\nwith a warning message. ´ZSTDNBTHREADShas a default value of (1), and is capped at ZST‐‐\nDMTNBWORKERSMAX==200.zstd` must be compiled with multithread support for this to have any\neffect.\n\nThey can both be overridden by corresponding command line arguments: -# for compression level\nand -T# for number of compression threads.\n\n### DICTIONARY BUILDER\n\nzstd offers dictionary compression, which greatly improves efficiency on small files and mes‐\nsages. It´s possible to train zstd with a set of samples, the result of which is saved into a\nfile called a dictionary. Then during compression and decompression, reference the same dic‐\ntionary, using command -D dictionaryFileName. Compression of small files similar to the sam‐\nple set will be greatly improved.\n\n#### --train FILEs\n\nUse FILEs as training set to create a dictionary. The training set should contain a\nlot of small files (> 100), and weight typically 100x the target dictionary size (for\nexample, 10 MB for a 100 KB dictionary).\n\nSupports multithreading if zstd is compiled with threading support. Additional parame‐\nters can be specified with --train-fastcover. The legacy dictionary builder can be ac‐\ncessed with --train-legacy. The cover dictionary builder can be accessed with\n--train-cover. Equivalent to --train-fastcover=d=8,steps=4.\n\n#### -o file\n\nDictionary saved into file (default name: dictionary).\n\n#### --maxdict=#\n\nLimit dictionary to specified size (default: 112640).\n\n-# Use # compression level during training (optional). Will generate statistics more\ntuned for selected compression level, resulting in a small compression ratio improve‐\nment for this level.\n\n#### -B#\n\n#### --dictID=#\n\nA dictionary ID is a locally unique ID that a decoder can use to verify it is using\nthe right dictionary. By default, zstd will create a 4-bytes random number ID. It´s\npossible to give a precise number instead. Short numbers have an advantage : an ID <\n256 will only need 1 byte in the compressed frame header, and an ID < 65536 will only\nneed 2 bytes. This compares favorably to 4 bytes default. However, it´s up to the dic‐\ntionary manager to not assign twice the same ID to 2 different dictionaries.\n\n#### --train-cover[=k#,d=#,steps=#,split=#,shrink[=#]]\n\nSelect parameters for the default dictionary builder algorithm named cover. If d is\nnot specified, then it tries d = 6 and d = 8. If k is not specified, then it tries\nsteps values in the range [50, 2000]. If steps is not specified, then the default\nvalue of 40 is used. If split is not specified or split <= 0, then the default value\nof 100 is used. Requires that d <= k. If shrink flag is not used, then the default\nvalue for shrinkDict of 0 is used. If shrink is not specified, then the default value\nfor shrinkDictMaxRegression of 1 is used.\n\nSelects segments of size k with highest score to put in the dictionary. The score of a\nsegment is computed by the sum of the frequencies of all the subsegments of size d.\nGenerally d should be in the range [6, 8], occasionally up to 16, but the algorithm\nwill run faster with d <= 8. Good values for k vary widely based on the input data,\nbut a safe range is [2 * d, 2000]. If split is 100, all input samples are used for\nboth training and testing to find optimal d and k to build dictionary. Supports multi‐\nthreading if zstd is compiled with threading support. Having shrink enabled takes a\ntruncated dictionary of minimum size and doubles in size until compression ratio of\nthe truncated dictionary is at most shrinkDictMaxRegression% worse than the compres‐\nsion ratio of the largest dictionary.\n\nExamples:\n\nzstd --train-cover FILEs\n\nzstd --train-cover=k=50,d=8 FILEs\n\nzstd --train-cover=d=8,steps=500 FILEs\n\nzstd --train-cover=k=50 FILEs\n\nzstd --train-cover=k=50,split=60 FILEs\n\nzstd --train-cover=shrink FILEs\n\nzstd --train-cover=shrink=2 FILEs\n\n#### --train-fastcover[=k#,d=#,f=#,steps=#,split=#,accel=#]\n\nSame as cover but with extra parameters f and accel and different default value of\nsplit If split is not specified, then it tries split = 75. If f is not specified, then\nit tries f = 20. Requires that 0 < f < 32. If accel is not specified, then it tries\naccel = 1. Requires that 0 < accel <= 10. Requires that d = 6 or d = 8.\n\nf is log of size of array that keeps track of frequency of subsegments of size d. The\nsubsegment is hashed to an index in the range [0,2^f - 1]. It is possible that 2 dif‐\nferent subsegments are hashed to the same index, and they are considered as the same\nsubsegment when computing frequency. Using a higher f reduces collision but takes\nlonger.\n\nExamples:\n\nzstd --train-fastcover FILEs\n\nzstd --train-fastcover=d=8,f=15,accel=2 FILEs\n\n#### --train-legacy[=selectivity=#]\n\nUse legacy dictionary builder algorithm with the given dictionary selectivity (de‐\nfault: 9). The smaller the selectivity value, the denser the dictionary, improving its\nefficiency but reducing its possible maximum size. --train-legacy=s=# is also ac‐\ncepted.\n\nExamples:\n\nzstd --train-legacy FILEs\n\nzstd --train-legacy=selectivity=8 FILEs\n\n### BENCHMARK\n\n#### -b#\n\n#### -e# -b# -e#\n\n#### -i#\n\n#### -B# --block-size=#\n\ncut file(s) into independent blocks of size # (default: no block)\n\n#### --priority=rt\n\nset process priority to real-time\n\nOutput Format: CompressionLevel#Filename : IntputSize -> OutputSize (CompressionRatio), Com‐\npressionSpeed, DecompressionSpeed\n\nMethodology: For both compression and decompression speed, the entire input is compressed/de‐\ncompressed in-memory to measure speed. A run lasts at least 1 sec, so when files are small,\nthey are compressed/decompressed several times per run, in order to improve measurement accu‐\nracy.\n\n### ADVANCED COMPRESSION OPTIONS\n\n#### --zstd[=options]:\n\nzstd provides 22 predefined compression levels. The selected or default predefined compres‐\nsion level can be changed with advanced compression options. The options are provided as a\ncomma-separated list. You may specify only the options you want to change and the rest will\nbe taken from the selected or default compression level. The list of available options:\n\nstrategy=strat, strat=strat\nSpecify a strategy used by a match finder.\n\nThere are 9 strategies numbered from 1 to 9, from faster to stronger: 1=ZSTDfast,\n2=ZSTDdfast, 3=ZSTDgreedy, 4=ZSTDlazy, 5=ZSTDlazy2, 6=ZSTDbtlazy2, 7=ZSTDbtopt,\n8=ZSTDbtultra, 9=ZSTDbtultra2.\n\nwindowLog=wlog, wlog=wlog\nSpecify the maximum number of bits for a match distance.\n\nThe higher number of increases the chance to find a match which usually improves com‐\npression ratio. It also increases memory requirements for the compressor and decom‐\npressor. The minimum wlog is 10 (1 KiB) and the maximum is 30 (1 GiB) on 32-bit plat‐\nforms and 31 (2 GiB) on 64-bit platforms.\n\nNote: If windowLog is set to larger than 27, --long=windowLog or --memory=windowSize\nneeds to be passed to the decompressor.\n\nhashLog=hlog, hlog=hlog\nSpecify the maximum number of bits for a hash table.\n\nBigger hash tables cause less collisions which usually makes compression faster, but\nrequires more memory during compression.\n\nThe minimum hlog is 6 (64 B) and the maximum is 30 (1 GiB).\n\nchainLog=clog, clog=clog\nSpecify the maximum number of bits for a hash chain or a binary tree.\n\nHigher numbers of bits increases the chance to find a match which usually improves\ncompression ratio. It also slows down compression speed and increases memory require‐\nments for compression. This option is ignored for the ZSTDfast strategy.\n\nThe minimum clog is 6 (64 B) and the maximum is 29 (524 Mib) on 32-bit platforms and\n30 (1 Gib) on 64-bit platforms.\n\nsearchLog=slog, slog=slog\nSpecify the maximum number of searches in a hash chain or a binary tree using loga‐\nrithmic scale.\n\nMore searches increases the chance to find a match which usually increases compression\nratio but decreases compression speed.\n\nThe minimum slog is 1 and the maximum is ´windowLog´ - 1.\n\nminMatch=mml, mml=mml\nSpecify the minimum searched length of a match in a hash table.\n\nLarger search lengths usually decrease compression ratio but improve decompression\nspeed.\n\nThe minimum mml is 3 and the maximum is 7.\n\ntargetLength=tlen, tlen=tlen\nThe impact of this field vary depending on selected strategy.\n\nFor ZSTDbtopt, ZSTDbtultra and ZSTDbtultra2, it specifies the minimum match length\nthat causes match finder to stop searching. A larger targetLength usually improves\ncompression ratio but decreases compression speed. t For ZSTDfast, it triggers ul‐\ntra-fast mode when > 0. The value represents the amount of data skipped between match\nsampling. Impact is reversed : a larger targetLength increases compression speed but\ndecreases compression ratio.\n\nFor all other strategies, this field has no impact.\n\nThe minimum tlen is 0 and the maximum is 128 Kib.\n\noverlapLog=ovlog, ovlog=ovlog\nDetermine overlapSize, amount of data reloaded from previous job. This parameter is\nonly available when multithreading is enabled. Reloading more data improves compres‐\nsion ratio, but decreases speed.\n\nThe minimum ovlog is 0, and the maximum is 9. 1 means \"no overlap\", hence completely\nindependent jobs. 9 means \"full overlap\", meaning up to windowSize is reloaded from\nprevious job. Reducing ovlog by 1 reduces the reloaded amount by a factor 2. For exam‐\nple, 8 means \"windowSize/2\", and 6 means \"windowSize/8\". Value 0 is special and means\n\"default\" : ovlog is automatically determined by zstd. In which case, ovlog will range\nfrom 6 to 9, depending on selected strat.\n\nldmHashLog=lhlog, lhlog=lhlog\nSpecify the maximum size for a hash table used for long distance matching.\n\nThis option is ignored unless long distance matching is enabled.\n\nBigger hash tables usually improve compression ratio at the expense of more memory\nduring compression and a decrease in compression speed.\n\nThe minimum lhlog is 6 and the maximum is 30 (default: 20).\n\nldmMinMatch=lmml, lmml=lmml\nSpecify the minimum searched length of a match for long distance matching.\n\nThis option is ignored unless long distance matching is enabled.\n\nLarger/very small values usually decrease compression ratio.\n\nThe minimum lmml is 4 and the maximum is 4096 (default: 64).\n\nldmBucketSizeLog=lblog, lblog=lblog\nSpecify the size of each bucket for the hash table used for long distance matching.\n\nThis option is ignored unless long distance matching is enabled.\n\nLarger bucket sizes improve collision resolution but decrease compression speed.\n\nThe minimum lblog is 1 and the maximum is 8 (default: 3).\n\nldmHashRateLog=lhrlog, lhrlog=lhrlog\nSpecify the frequency of inserting entries into the long distance matching hash table.\n\nThis option is ignored unless long distance matching is enabled.\n\nLarger values will improve compression speed. Deviating far from the default value\nwill likely result in a decrease in compression ratio.\n\nThe default value is wlog - lhlog.\n\n#### Example\n\nThe following parameters sets advanced compression options to something similar to predefined\nlevel 19 for files bigger than 256 KB:\n\n--zstd=wlog=23,clog=23,hlog=22,slog=6,mml=3,tlen=48,strat=6\n\n#### -B#:\n\nSelect the size of each compression job. This parameter is available only when multi-thread‐\ning is enabled. Default value is 4 * windowSize, which means it varies depending on compres‐\nsion level. -B# makes it possible to select a custom value. Note that job size must respect a\nminimum value which is enforced transparently. This minimum is either 1 MB, or overlapSize,\nwhichever is largest.\n\n### BUGS\n\nReport bugs at: https://github.com/facebook/zstd/issues\n\n### AUTHOR\n\nYann Collet\n\n\n\nzstd 1.4.8 December 2020 ZSTD(1)\n\n" } ], "structuredContent": { "command": "unzstd", "section": "1", "mode": "man", "summary": "zstd - zstd, zstdmt, unzstd, zstdcat - Compress or decompress .zst files", "synopsis": "zstd [OPTIONS] [-|INPUT-FILE] [-o OUTPUT-FILE]\nzstdmt is equivalent to zstd -T0\nunzstd is equivalent to zstd -d\nzstdcat is equivalent to zstd -dcf", "flags": [ { "flag": "-z", "long": "--compress", "arg": null, "description": "Compress. This is the default operation mode when no operation mode option is speci‐ fied and no other operation mode is implied from the command name (for example, unzstd implies --decompress)." }, { "flag": "-d", "long": "--uncompress", "arg": null, "description": "Decompress." }, { "flag": "-t", "long": "--test", "arg": null, "description": "Test the integrity of compressed files. This option is equivalent to --decompress --stdout except that the decompressed data is discarded instead of being written to standard output. No files are created or removed." }, { "flag": "", "long": null, "arg": null, "description": "" }, { "flag": "", "long": "--train", "arg": null, "description": "Use FILEs as a training set to create a dictionary. The training set should contain a lot of small files (> 100)." }, { "flag": "-l", "long": "--list", "arg": null, "description": "Display information related to a zstd compressed file, such as size, ratio, and check‐ sum. Some of these fields may not be available. This command can be augmented with the -v modifier." } ], "examples": [], "see_also": [], "section_outline": [ { "name": "NAME", "lines": 2, "subsections": [] }, { "name": "SYNOPSIS", "lines": 8, "subsections": [] }, { "name": "DESCRIPTION", "lines": 39, "subsections": [ { "name": "Concatenation with .zst files", "lines": 3 } ] }, { "name": "OPTIONS", "lines": 1, "subsections": [ { "name": "Integer suffixes and special values", "lines": 8 }, { "name": "Operation mode", "lines": 2 }, { "name": "-z --compress", "lines": 4, "flag": "-z", "long": "--compress" }, { "name": "-d --decompress --uncompress", "lines": 2, "flag": "-d", "long": "--uncompress" }, { "name": "-t --test", "lines": 4, "flag": "-t", "long": "--test" }, { "name": "-b#", "lines": 1 }, { "name": "--train FILEs", "lines": 3, "long": "--train" }, { "name": "-l --list", "lines": 4, "flag": "-l", "long": "--list" }, { "name": "Operation modifiers", "lines": 156 }, { "name": "Restricted usage of Environment Variables", "lines": 18 } ] }, { "name": "DICTIONARY BUILDER", "lines": 6, "subsections": [ { "name": "--train FILEs", "lines": 9, "long": "--train" }, { "name": "-o file", "lines": 2, "flag": "-o" }, { "name": "--maxdict=#", "lines": 6, "long": "--maxdict", "arg": "#" }, { "name": "-B#", "lines": 1 }, { "name": "--dictID=#", "lines": 7, "long": "--dictID", "arg": "#" }, { "name": "--train-cover[=k#,d=#,steps=#,split=#,shrink[=#]]", "lines": 35 }, { "name": "--train-fastcover[=k#,d=#,f=#,steps=#,split=#,accel=#]", "lines": 17 }, { "name": "--train-legacy[=selectivity=#]", "lines": 11 } ] }, { "name": "BENCHMARK", "lines": 1, "subsections": [ { "name": "-b#", "lines": 1 }, { "name": "-e# -b# -e#", "lines": 1 }, { "name": "-i#", "lines": 1 }, { "name": "-B# --block-size=#", "lines": 2, "long": "--block-size", "arg": "#" }, { "name": "--priority=rt", "lines": 10, "long": "--priority", "arg": "rt" } ] }, { "name": "ADVANCED COMPRESSION OPTIONS", "lines": 1, "subsections": [ { "name": "--zstd[=options]:", "lines": 122 }, { "name": "Example", "lines": 5 }, { "name": "-B#:", "lines": 6 } ] }, { "name": "BUGS", "lines": 2, "subsections": [] }, { "name": "AUTHOR", "lines": 5, "subsections": [] } ] } }