zp-1.0.orig/0002755000000000000500000000000011403164663011257 5ustar rootsrczp-1.0.orig/LICENSE0000644000000000000500000010476611376275540012306 0ustar rootsrcLICENSE description added by Matt Mahoney on May 23, 2010. All code with the exception of the SHA1 class is Copyright (C), Ocarina Networks Inc, as dated in the source code, and is licensed under the GNU General Public License, version 3. The SHA1 class is derived from code in RFC-3174, which is Copyright (C), 2001, The Internet Society. Both licenses are included below. -------------------------------------------------------------------------- License for code derived from RFC-3174 (class SHA1). Source: http://datatracker.ietf.org/doc/rfc3174/ Copyright (C) The Internet Society (2001). All Rights Reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Internet organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by the Internet Society or its successors or assigns. This document and the information contained herein is provided on an "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. -------------------------------------------------------------------------- License for all code not derived from RFC-3174. Source: http://www.gnu.org/licenses/gpl.txt GNU GENERAL PUBLIC LICENSE Version 3, 29 June 2007 Copyright (C) 2007 Free Software Foundation, Inc. Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. Preamble The GNU General Public License is a free, copyleft license for software and other kinds of works. The licenses for most software and other practical works are designed to take away your freedom to share and change the works. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change all versions of a program--to make sure it remains free software for all its users. We, the Free Software Foundation, use the GNU General Public License for most of our software; it applies also to any other work released this way by its authors. You can apply it to your programs, too. When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for them if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs, and that you know you can do these things. To protect your rights, we need to prevent others from denying you these rights or asking you to surrender the rights. Therefore, you have certain responsibilities if you distribute copies of the software, or if you modify it: responsibilities to respect the freedom of others. For example, if you distribute copies of such a program, whether gratis or for a fee, you must pass on to the recipients the same freedoms that you received. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights. Developers that use the GNU GPL protect your rights with two steps: (1) assert copyright on the software, and (2) offer you this License giving you legal permission to copy, distribute and/or modify it. For the developers' and authors' protection, the GPL clearly explains that there is no warranty for this free software. For both users' and authors' sake, the GPL requires that modified versions be marked as changed, so that their problems will not be attributed erroneously to authors of previous versions. Some devices are designed to deny users access to install or run modified versions of the software inside them, although the manufacturer can do so. This is fundamentally incompatible with the aim of protecting users' freedom to change the software. The systematic pattern of such abuse occurs in the area of products for individuals to use, which is precisely where it is most unacceptable. Therefore, we have designed this version of the GPL to prohibit the practice for those products. If such problems arise substantially in other domains, we stand ready to extend this provision to those domains in future versions of the GPL, as needed to protect the freedom of users. Finally, every program is threatened constantly by software patents. States should not allow patents to restrict development and use of software on general-purpose computers, but in those that do, we wish to avoid the special danger that patents applied to a free program could make it effectively proprietary. To prevent this, the GPL assures that patents cannot be used to render the program non-free. The precise terms and conditions for copying, distribution and modification follow. TERMS AND CONDITIONS 0. Definitions. "This License" refers to version 3 of the GNU General Public License. "Copyright" also means copyright-like laws that apply to other kinds of works, such as semiconductor masks. "The Program" refers to any copyrightable work licensed under this License. Each licensee is addressed as "you". "Licensees" and "recipients" may be individuals or organizations. To "modify" a work means to copy from or adapt all or part of the work in a fashion requiring copyright permission, other than the making of an exact copy. The resulting work is called a "modified version" of the earlier work or a work "based on" the earlier work. A "covered work" means either the unmodified Program or a work based on the Program. To "propagate" a work means to do anything with it that, without permission, would make you directly or secondarily liable for infringement under applicable copyright law, except executing it on a computer or modifying a private copy. Propagation includes copying, distribution (with or without modification), making available to the public, and in some countries other activities as well. To "convey" a work means any kind of propagation that enables other parties to make or receive copies. Mere interaction with a user through a computer network, with no transfer of a copy, is not conveying. An interactive user interface displays "Appropriate Legal Notices" to the extent that it includes a convenient and prominently visible feature that (1) displays an appropriate copyright notice, and (2) tells the user that there is no warranty for the work (except to the extent that warranties are provided), that licensees may convey the work under this License, and how to view a copy of this License. If the interface presents a list of user commands or options, such as a menu, a prominent item in the list meets this criterion. 1. Source Code. The "source code" for a work means the preferred form of the work for making modifications to it. "Object code" means any non-source form of a work. A "Standard Interface" means an interface that either is an official standard defined by a recognized standards body, or, in the case of interfaces specified for a particular programming language, one that is widely used among developers working in that language. The "System Libraries" of an executable work include anything, other than the work as a whole, that (a) is included in the normal form of packaging a Major Component, but which is not part of that Major Component, and (b) serves only to enable use of the work with that Major Component, or to implement a Standard Interface for which an implementation is available to the public in source code form. A "Major Component", in this context, means a major essential component (kernel, window system, and so on) of the specific operating system (if any) on which the executable work runs, or a compiler used to produce the work, or an object code interpreter used to run it. The "Corresponding Source" for a work in object code form means all the source code needed to generate, install, and (for an executable work) run the object code and to modify the work, including scripts to control those activities. However, it does not include the work's System Libraries, or general-purpose tools or generally available free programs which are used unmodified in performing those activities but which are not part of the work. For example, Corresponding Source includes interface definition files associated with source files for the work, and the source code for shared libraries and dynamically linked subprograms that the work is specifically designed to require, such as by intimate data communication or control flow between those subprograms and other parts of the work. The Corresponding Source need not include anything that users can regenerate automatically from other parts of the Corresponding Source. The Corresponding Source for a work in source code form is that same work. 2. Basic Permissions. All rights granted under this License are granted for the term of copyright on the Program, and are irrevocable provided the stated conditions are met. This License explicitly affirms your unlimited permission to run the unmodified Program. The output from running a covered work is covered by this License only if the output, given its content, constitutes a covered work. This License acknowledges your rights of fair use or other equivalent, as provided by copyright law. You may make, run and propagate covered works that you do not convey, without conditions so long as your license otherwise remains in force. You may convey covered works to others for the sole purpose of having them make modifications exclusively for you, or provide you with facilities for running those works, provided that you comply with the terms of this License in conveying all material for which you do not control copyright. Those thus making or running the covered works for you must do so exclusively on your behalf, under your direction and control, on terms that prohibit them from making any copies of your copyrighted material outside their relationship with you. Conveying under any other circumstances is permitted solely under the conditions stated below. Sublicensing is not allowed; section 10 makes it unnecessary. 3. Protecting Users' Legal Rights From Anti-Circumvention Law. No covered work shall be deemed part of an effective technological measure under any applicable law fulfilling obligations under article 11 of the WIPO copyright treaty adopted on 20 December 1996, or similar laws prohibiting or restricting circumvention of such measures. When you convey a covered work, you waive any legal power to forbid circumvention of technological measures to the extent such circumvention is effected by exercising rights under this License with respect to the covered work, and you disclaim any intention to limit operation or modification of the work as a means of enforcing, against the work's users, your or third parties' legal rights to forbid circumvention of technological measures. 4. Conveying Verbatim Copies. You may convey verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice; keep intact all notices stating that this License and any non-permissive terms added in accord with section 7 apply to the code; keep intact all notices of the absence of any warranty; and give all recipients a copy of this License along with the Program. You may charge any price or no price for each copy that you convey, and you may offer support or warranty protection for a fee. 5. Conveying Modified Source Versions. You may convey a work based on the Program, or the modifications to produce it from the Program, in the form of source code under the terms of section 4, provided that you also meet all of these conditions: a) The work must carry prominent notices stating that you modified it, and giving a relevant date. b) The work must carry prominent notices stating that it is released under this License and any conditions added under section 7. This requirement modifies the requirement in section 4 to "keep intact all notices". c) You must license the entire work, as a whole, under this License to anyone who comes into possession of a copy. This License will therefore apply, along with any applicable section 7 additional terms, to the whole of the work, and all its parts, regardless of how they are packaged. This License gives no permission to license the work in any other way, but it does not invalidate such permission if you have separately received it. d) If the work has interactive user interfaces, each must display Appropriate Legal Notices; however, if the Program has interactive interfaces that do not display Appropriate Legal Notices, your work need not make them do so. A compilation of a covered work with other separate and independent works, which are not by their nature extensions of the covered work, and which are not combined with it such as to form a larger program, in or on a volume of a storage or distribution medium, is called an "aggregate" if the compilation and its resulting copyright are not used to limit the access or legal rights of the compilation's users beyond what the individual works permit. Inclusion of a covered work in an aggregate does not cause this License to apply to the other parts of the aggregate. 6. Conveying Non-Source Forms. You may convey a covered work in object code form under the terms of sections 4 and 5, provided that you also convey the machine-readable Corresponding Source under the terms of this License, in one of these ways: a) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by the Corresponding Source fixed on a durable physical medium customarily used for software interchange. b) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by a written offer, valid for at least three years and valid for as long as you offer spare parts or customer support for that product model, to give anyone who possesses the object code either (1) a copy of the Corresponding Source for all the software in the product that is covered by this License, on a durable physical medium customarily used for software interchange, for a price no more than your reasonable cost of physically performing this conveying of source, or (2) access to copy the Corresponding Source from a network server at no charge. c) Convey individual copies of the object code with a copy of the written offer to provide the Corresponding Source. This alternative is allowed only occasionally and noncommercially, and only if you received the object code with such an offer, in accord with subsection 6b. d) Convey the object code by offering access from a designated place (gratis or for a charge), and offer equivalent access to the Corresponding Source in the same way through the same place at no further charge. You need not require recipients to copy the Corresponding Source along with the object code. If the place to copy the object code is a network server, the Corresponding Source may be on a different server (operated by you or a third party) that supports equivalent copying facilities, provided you maintain clear directions next to the object code saying where to find the Corresponding Source. Regardless of what server hosts the Corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements. e) Convey the object code using peer-to-peer transmission, provided you inform other peers where the object code and Corresponding Source of the work are being offered to the general public at no charge under subsection 6d. A separable portion of the object code, whose source code is excluded from the Corresponding Source as a System Library, need not be included in conveying the object code work. A "User Product" is either (1) a "consumer product", which means any tangible personal property which is normally used for personal, family, or household purposes, or (2) anything designed or sold for incorporation into a dwelling. In determining whether a product is a consumer product, doubtful cases shall be resolved in favor of coverage. For a particular product received by a particular user, "normally used" refers to a typical or common use of that class of product, regardless of the status of the particular user or of the way in which the particular user actually uses, or expects or is expected to use, the product. A product is a consumer product regardless of whether the product has substantial commercial, industrial or non-consumer uses, unless such uses represent the only significant mode of use of the product. "Installation Information" for a User Product means any methods, procedures, authorization keys, or other information required to install and execute modified versions of a covered work in that User Product from a modified version of its Corresponding Source. The information must suffice to ensure that the continued functioning of the modified object code is in no case prevented or interfered with solely because modification has been made. If you convey an object code work under this section in, or with, or specifically for use in, a User Product, and the conveying occurs as part of a transaction in which the right of possession and use of the User Product is transferred to the recipient in perpetuity or for a fixed term (regardless of how the transaction is characterized), the Corresponding Source conveyed under this section must be accompanied by the Installation Information. But this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product (for example, the work has been installed in ROM). The requirement to provide Installation Information does not include a requirement to continue to provide support service, warranty, or updates for a work that has been modified or installed by the recipient, or for the User Product in which it has been modified or installed. Access to a network may be denied when the modification itself materially and adversely affects the operation of the network or violates the rules and protocols for communication across the network. Corresponding Source conveyed, and Installation Information provided, in accord with this section must be in a format that is publicly documented (and with an implementation available to the public in source code form), and must require no special password or key for unpacking, reading or copying. 7. Additional Terms. "Additional permissions" are terms that supplement the terms of this License by making exceptions from one or more of its conditions. Additional permissions that are applicable to the entire Program shall be treated as though they were included in this License, to the extent that they are valid under applicable law. If additional permissions apply only to part of the Program, that part may be used separately under those permissions, but the entire Program remains governed by this License without regard to the additional permissions. When you convey a copy of a covered work, you may at your option remove any additional permissions from that copy, or from any part of it. (Additional permissions may be written to require their own removal in certain cases when you modify the work.) You may place additional permissions on material, added by you to a covered work, for which you have or can give appropriate copyright permission. Notwithstanding any other provision of this License, for material you add to a covered work, you may (if authorized by the copyright holders of that material) supplement the terms of this License with terms: a) Disclaiming warranty or limiting liability differently from the terms of sections 15 and 16 of this License; or b) Requiring preservation of specified reasonable legal notices or author attributions in that material or in the Appropriate Legal Notices displayed by works containing it; or c) Prohibiting misrepresentation of the origin of that material, or requiring that modified versions of such material be marked in reasonable ways as different from the original version; or d) Limiting the use for publicity purposes of names of licensors or authors of the material; or e) Declining to grant rights under trademark law for use of some trade names, trademarks, or service marks; or f) Requiring indemnification of licensors and authors of that material by anyone who conveys the material (or modified versions of it) with contractual assumptions of liability to the recipient, for any liability that these contractual assumptions directly impose on those licensors and authors. All other non-permissive additional terms are considered "further restrictions" within the meaning of section 10. If the Program as you received it, or any part of it, contains a notice stating that it is governed by this License along with a term that is a further restriction, you may remove that term. If a license document contains a further restriction but permits relicensing or conveying under this License, you may add to a covered work material governed by the terms of that license document, provided that the further restriction does not survive such relicensing or conveying. If you add terms to a covered work in accord with this section, you must place, in the relevant source files, a statement of the additional terms that apply to those files, or a notice indicating where to find the applicable terms. Additional terms, permissive or non-permissive, may be stated in the form of a separately written license, or stated as exceptions; the above requirements apply either way. 8. Termination. You may not propagate or modify a covered work except as expressly provided under this License. Any attempt otherwise to propagate or modify it is void, and will automatically terminate your rights under this License (including any patent licenses granted under the third paragraph of section 11). However, if you cease all violation of this License, then your license from a particular copyright holder is reinstated (a) provisionally, unless and until the copyright holder explicitly and finally terminates your license, and (b) permanently, if the copyright holder fails to notify you of the violation by some reasonable means prior to 60 days after the cessation. Moreover, your license from a particular copyright holder is reinstated permanently if the copyright holder notifies you of the violation by some reasonable means, this is the first time you have received notice of violation of this License (for any work) from that copyright holder, and you cure the violation prior to 30 days after your receipt of the notice. Termination of your rights under this section does not terminate the licenses of parties who have received copies or rights from you under this License. If your rights have been terminated and not permanently reinstated, you do not qualify to receive new licenses for the same material under section 10. 9. Acceptance Not Required for Having Copies. You are not required to accept this License in order to receive or run a copy of the Program. Ancillary propagation of a covered work occurring solely as a consequence of using peer-to-peer transmission to receive a copy likewise does not require acceptance. However, nothing other than this License grants you permission to propagate or modify any covered work. These actions infringe copyright if you do not accept this License. Therefore, by modifying or propagating a covered work, you indicate your acceptance of this License to do so. 10. Automatic Licensing of Downstream Recipients. Each time you convey a covered work, the recipient automatically receives a license from the original licensors, to run, modify and propagate that work, subject to this License. You are not responsible for enforcing compliance by third parties with this License. An "entity transaction" is a transaction transferring control of an organization, or substantially all assets of one, or subdividing an organization, or merging organizations. If propagation of a covered work results from an entity transaction, each party to that transaction who receives a copy of the work also receives whatever licenses to the work the party's predecessor in interest had or could give under the previous paragraph, plus a right to possession of the Corresponding Source of the work from the predecessor in interest, if the predecessor has it or can get it with reasonable efforts. You may not impose any further restrictions on the exercise of the rights granted or affirmed under this License. For example, you may not impose a license fee, royalty, or other charge for exercise of rights granted under this License, and you may not initiate litigation (including a cross-claim or counterclaim in a lawsuit) alleging that any patent claim is infringed by making, using, selling, offering for sale, or importing the Program or any portion of it. 11. Patents. A "contributor" is a copyright holder who authorizes use under this License of the Program or a work on which the Program is based. The work thus licensed is called the contributor's "contributor version". A contributor's "essential patent claims" are all patent claims owned or controlled by the contributor, whether already acquired or hereafter acquired, that would be infringed by some manner, permitted by this License, of making, using, or selling its contributor version, but do not include claims that would be infringed only as a consequence of further modification of the contributor version. For purposes of this definition, "control" includes the right to grant patent sublicenses in a manner consistent with the requirements of this License. Each contributor grants you a non-exclusive, worldwide, royalty-free patent license under the contributor's essential patent claims, to make, use, sell, offer for sale, import and otherwise run, modify and propagate the contents of its contributor version. In the following three paragraphs, a "patent license" is any express agreement or commitment, however denominated, not to enforce a patent (such as an express permission to practice a patent or covenant not to sue for patent infringement). To "grant" such a patent license to a party means to make such an agreement or commitment not to enforce a patent against the party. If you convey a covered work, knowingly relying on a patent license, and the Corresponding Source of the work is not available for anyone to copy, free of charge and under the terms of this License, through a publicly available network server or other readily accessible means, then you must either (1) cause the Corresponding Source to be so available, or (2) arrange to deprive yourself of the benefit of the patent license for this particular work, or (3) arrange, in a manner consistent with the requirements of this License, to extend the patent license to downstream recipients. "Knowingly relying" means you have actual knowledge that, but for the patent license, your conveying the covered work in a country, or your recipient's use of the covered work in a country, would infringe one or more identifiable patents in that country that you have reason to believe are valid. If, pursuant to or in connection with a single transaction or arrangement, you convey, or propagate by procuring conveyance of, a covered work, and grant a patent license to some of the parties receiving the covered work authorizing them to use, propagate, modify or convey a specific copy of the covered work, then the patent license you grant is automatically extended to all recipients of the covered work and works based on it. A patent license is "discriminatory" if it does not include within the scope of its coverage, prohibits the exercise of, or is conditioned on the non-exercise of one or more of the rights that are specifically granted under this License. You may not convey a covered work if you are a party to an arrangement with a third party that is in the business of distributing software, under which you make payment to the third party based on the extent of your activity of conveying the work, and under which the third party grants, to any of the parties who would receive the covered work from you, a discriminatory patent license (a) in connection with copies of the covered work conveyed by you (or copies made from those copies), or (b) primarily for and in connection with specific products or compilations that contain the covered work, unless you entered into that arrangement, or that patent license was granted, prior to 28 March 2007. Nothing in this License shall be construed as excluding or limiting any implied license or other defenses to infringement that may otherwise be available to you under applicable patent law. 12. No Surrender of Others' Freedom. If conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot convey a covered work so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not convey it at all. For example, if you agree to terms that obligate you to collect a royalty for further conveying from those to whom you convey the Program, the only way you could satisfy both those terms and this License would be to refrain entirely from conveying the Program. 13. Use with the GNU Affero General Public License. Notwithstanding any other provision of this License, you have permission to link or combine any covered work with a work licensed under version 3 of the GNU Affero General Public License into a single combined work, and to convey the resulting work. The terms of this License will continue to apply to the part which is the covered work, but the special requirements of the GNU Affero General Public License, section 13, concerning interaction through a network will apply to the combination as such. 14. Revised Versions of this License. The Free Software Foundation may publish revised and/or new versions of the GNU General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies that a certain numbered version of the GNU General Public License "or any later version" applies to it, you have the option of following the terms and conditions either of that numbered version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of the GNU General Public License, you may choose any version ever published by the Free Software Foundation. If the Program specifies that a proxy can decide which future versions of the GNU General Public License can be used, that proxy's public statement of acceptance of a version permanently authorizes you to choose that version for the Program. Later license versions may give you additional or different permissions. However, no additional obligations are imposed on any author or copyright holder as a result of your choosing to follow a later version. 15. Disclaimer of Warranty. THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 16. Limitation of Liability. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. 17. Interpretation of Sections 15 and 16. If the disclaimer of warranty and limitation of liability provided above cannot be given local legal effect according to their terms, reviewing courts shall apply local law that most closely approximates an absolute waiver of all civil liability in connection with the Program, unless a warranty or assumption of liability accompanies a copy of the Program in return for a fee. END OF TERMS AND CONDITIONS zp-1.0.orig/zp.cpp0000644000000000000500000031203311365442442012415 0ustar rootsrc/* zp v1.00 archiver and file compressor. Written by Matt Mahoney, matmahoney@yahoo.com, Apr. 26, 2010. Copyright (C) 2010, Ocarina Networks, Inc. LICENSE This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 3 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details at Visit . Usage: zp command archive.zpaq files... Commands: l - list contents of archive.zpaq x - extract with full path names (files... overrides stored names) e - extract to current directory xN, eN - extract block N only (starting with 1) cN - create new archive with compression option N aN - append to archive with option N Compression options N are 1=fast, 2=medium (default), 3=small. Archives created with zp conform to the ZPAQ level 1 standard described at http://mattmahoney.net/dc/ Archives are read/write compatible with other compliant programs such as zpaq, unzpaq, and zpipe. Recommended options for Windows/g++ 4.4.0: g++ -O2 -s -march=pentiumpro -fomit-frame-pointer zp.cpp -o zp.exe upx zp.exe Compile with -Dunix in Linux. (Usually this is automatic). Compile with -DDEBUG to enable run time checks. Command details: The archive name must end with ".zpaq". All commands will add the extension automatically if you don't specify it. For example: zp c3 arc file1 file2 zp a1 arc file3 will create archive arc.zpaq, compress file1 and file2 with smallest (slowest) compression, and append file3 with the fastest (least) compression. The commands "c" and "a" are equivalent to "c2" and "a2" (medium compression). The files are grouped into one block (solid archive) for each command. zp l arc will show the contents of arc.zpaq. It will show that file1 and file2 are stored in block 1, and file3 in block 2. zp x arc will extract file1, file2, and file3. You can extract from just one block: zp x1 arc will extract file1 and file2 only. zp x2 arc will extract file3 only. If you specify file names on the command line then the output files will be renamed in the order they are listed and extracted. zp x arc newfile1 will extract file1 as newfile1. It will not extract file2 or file3. zp x2 arc newfile3 will extract the first file of block 2 (file3) as newfile3. Blocks are "solid" which means you cannot extract files within a block without extracting the earlier files. For example, you cannot extract file2 without also extracting file1. zp will not clobber existing files during extraction unless you specify the filenames on the command line. zp x arc (Error: file1 exists) zp x arc file1 file2 file3 (Overwrites file1, file2, file3) File names are stored in the archive as they appear on the command line. If you specify a path to a different directory, the path is stored, and created during extraction. The "e" command extracts to the current directory. zp c arc dir1\dir2\file1 zp x arc will create dir1 and dir1\dir2 in the current directory if they do not already exist, then create dir1\dir2\file1 zp e arc will create file1 in the current directory (unless it exists). If you specify the output filenames, then "e" behaves the same as "x". If you compress in Windows and extract in Linux, then the program will change "\" to "/" during extraction and vice versa. Slashes can be stored with either convention. (The program guesses the operating system by counting "/" and "\" in the PATH environment variable. If this heuristic fails (PATH not defined) then no slash translation is done). Paths must be relative to the current directory. The program will warn if you store an absolute path. You can only extract such files with "e" or by overriding the filename. zp c arc \dir1\dir2\file1 (Warning: starts with "\") zp x arc (Error: bad filename) zp e arc (OK: extracts file1 to current directory) zp x arc newfile (OK: extracts newfile to current directory) zp x arc \dir3\dir4\newfile (OK: creates \dir3\dir4 if needed) Also, the same rule applies to file names containing control characters, or longer than 511 characters, or that start with a drive letter like "C:" or that go up directories (contain ../ or ..\). If this program is run in Linux or UNIX or compiled with g++ in Windows then it will interpret wildcards on the command line in the usual way. A * matches any string and ? matches any character. zp c arc * will compress all files in the current directory to arc.zpaq. However, it will not recurse directories. You need to specify the files in each directory that you want to add. The program does not save file timestamps or permissions like some other archivers do. Extracted files are dated from the time of extraction with default permissions. If you need these capabilities, then create a tar file and compress that instead. The compression option 1, 2, or 3 means compress fast, medium, or small respectively. Better compression requires more time and memory. Decompression speed and memory are the same as for compression. Speed (T3200, 2.0 GHz) and memory usage are as follows. zip -9 compression is shown for comparison. All modes compress better (but slower) than zip. Memory Speed Calgary corpus ------ ----------- --------------- 1 (fast) 38 MB 0.7 sec/MB 807,214 bytes 2 (default) 111 MB 2.3 sec/MB 699,586 bytes 3 (small) 246 MB 6.4 sec/MB 644,545 bytes zip -9 <1 MB 0.13 sec/MB 1,020,719 bytes Options 1, 2, 3 are equivalent to fast.cfg, mid.cfg, and max.cfg respectively. For example, "zp c3 arc file" is equivalent to "zpaq ocmax.cfg arc.zpaq file". mid.cfg and max.cfg are the same as in the ZPAQ 1.10 distribution. (There is also a min.cfg which is different from fast.cfg. This program used compiled ZPAQL (generated by "zpaq oc") to compress and extract in each of the 3 modes about twice as fast as using interpreted code. It automatically recognizes these configurations even if they are produced by other programs. The default compression is the same as the default produced by zpaq and zpipe. If another program produces a different configuration, then this program will still correctly decompress it by interpreting the code, which is slower. Also, zpaq, unzpaq, and zpipe can decompress archives produced by this program. The config files are as follows (with $1 defaulted to 0). See the ZPAQ standard and ZPAQ 1.10 source code comments to interpret these configuation files. fast.cfg uses an order 2 ICM (indirect context model) and order 4 ISSE (indirect secondary symbol estimation) with no mixer. mid.cfg uses an order 0..5 ICM/ISSE chain, an order 7 match model and an order 1 mixer. max.cfg uses an order 0..5, 7 ICM/ISSE chain, order 8 mixer, models for text (order 0 and 1 words), sparse models (order 0 with gaps of 1, 2, 3), CCITT images, 2 parallel mixers (order 0 and 1), and 2 serial SSE stages (orders 0 and 1) with adaptive bypass. (fast.cfg (c1, a1)) comp 1 2 0 0 2 (hh hm ph pm n) 0 icm 16 (order 2) 1 isse 19 0 (order 4) hcomp *b=a a=0 (save in rotating buffer M) d=0 hash b-- hash *d=a d++ b-- hash b-- hash *d=a halt post 0 end (mid.cfg (c2, a2)) comp 3 3 0 0 8 (hh hm ph pm n) 0 icm 5 (order 0...5 chain) 1 isse 13 0 2 isse $1+17 1 3 isse $1+18 2 4 isse $1+18 3 5 isse $1+19 4 6 match $1+22 $1+24 (order 7) 7 mix 16 0 7 24 255 (order 1) hcomp c++ *c=a b=c a=0 (save in rotating buffer M) d= 1 hash *d=a (orders 1...5 for isse) b-- d++ hash *d=a b-- d++ hash *d=a b-- d++ hash *d=a b-- d++ hash *d=a b-- d++ hash b-- hash *d=a (order 7 for match) d++ a=*c a<<= 8 *d=a (order 1 for mix) halt post 0 end (max.cfg (c3, a3)) comp 5 9 0 0 22 (hh hm ph pm n) 0 const 160 1 icm 5 (orders 0-6) 2 isse 13 1 (sizebits j) 3 isse $1+16 2 4 isse $1+18 3 5 isse $1+19 4 6 isse $1+19 5 7 isse $1+20 6 8 match $1+22 $1+24 9 icm $1+17 (order 0 word) 10 isse $1+19 9 (order 1 word) 11 icm 13 (sparse with gaps 1-3) 12 icm 13 13 icm 13 14 icm 14 (pic) 15 mix 16 0 15 24 255 (mix orders 1 and 0) 16 mix 8 0 16 10 255 (including last mixer) 17 mix2 0 15 16 24 0 18 sse 8 17 32 255 (order 0) 19 mix2 8 17 18 16 255 20 sse 16 19 32 255 (order 1) 21 mix2 0 19 20 16 0 hcomp c++ *c=a b=c a=0 (save in rotating buffer) d= 2 hash *d=a b-- (orders 1,2,3,4,5,7) d++ hash *d=a b-- d++ hash *d=a b-- d++ hash *d=a b-- d++ hash *d=a b-- d++ hash b-- hash *d=a b-- d++ hash *d=a b-- (match, order 8) d++ a=*c a&~ 32 (lowercase words) a> 64 if a< 91 if (if a-z) d++ hashd d-- (update order 1 word hash) *d<>a a+=*d a*= 20 *d=a (order 0 word hash) jmp 9 endif endif (else not a letter) a=*d a== 0 ifnot (move word order 0 to 1) d++ *d=a d-- endif *d=0 (clear order 0 word hash) (end else) d++ d++ b=c b-- a=0 hash *d=a (sparse 2) d++ b-- a=0 hash *d=a (sparse 3) d++ b-- a=0 hash *d=a (sparse 4) d++ a=b a-= 212 b=a a=0 hash *d=a b<>a a-= 216 b<>a a=*b a&= 60 hashd (pic) d++ a=*c a<<= 9 *d=a (mix) d++ d++ d++ d++ d++ *d=a (sse) halt post 0 end This program stores a filename, comment, and SHA-1 checksum for each file. Other programs may omit these, but this program will still be able to decompress them. This program follows the convention that if the name is omitted, then the contents should be appended to the previous file. If the first filename is omitted, then you must supply it on the command line during extraction. Each filename on the command line replaces one named file in the archive. The comment is the original file size as a decimal string (exact to 2^52, over 4000 TB). */ #ifndef DEBUG // compile with -DDEBUG to enable debugging #define NDEBUG #endif #include #include #include #include #include #include #include #include #ifdef unix #include #include #include #endif const int LEVEL=1; // ZPAQ level 0=experimental 1=final // 1, 2, 4 byte unsigned integers typedef unsigned char U8; typedef unsigned short U16; typedef unsigned int U32; // Print an error message and exit static void error(const char* msg="") { fprintf(stderr, "\nError: %s\n", msg); exit(1); } // An Array of T is cleared and aligned on a 64 byte address // with no constructors called. No copy or assignment. // Array a(n, ex=0); - creates n< class Array { private: T *data; // user location of [0] on a 64 byte boundary int n; // user size-1 int offset; // distance back in bytes to start of actual allocation void operator=(const Array&); // no assignment Array(const Array&); // no copy public: Array(int sz=0, int ex=0): data(0), n(-1), offset(0) { resize(sz, ex);} // [0..sz-1] = 0 void resize(int sz, int ex=0); // change size, erase content to zeros ~Array() {resize(0);} // free memory int size() const {return n+1;} // get size T& operator[](int i) {assert(n>=0 && i>=0 && U32(i)<=U32(n)); return data[i];} T& operator()(int i) {assert(n>=0 && (n&(n+1))==0); return data[i&n];} }; // Change size to sz< void Array::resize(int sz, int ex) { while (ex>0) { if (sz<0 || sz>=(1<<30)) error("Array too big"); sz*=2, --ex; } if (sz<0) error("Array too big"); if (n>-1) { assert(offset>0 && offset<=64); assert((char*)data-offset); free((char*)data-offset); } n=-1; if (sz<=0) return; n=sz-1; data=(T*)calloc(64+(n+1)*sizeof(T), 1); if (!data) error("Out of memory"); offset=64-int((long)data&63); assert(offset>0 && offset<=64); data=(T*)((char*)data+offset); } // A Reader reads from a file or an array U8 p[n] class Reader { FILE *in; const U8 *ptr; int len; public: Reader(FILE *f): in(f), ptr(0), len(0) {} // Read from file Reader(const U8 *p, int n): in(0), ptr(p), len(n) {} // Read from p[n] int get() { // return 1 byte or EOF if (in) return getc(in); else if (ptr && len) return --len, *ptr++; return EOF; } }; // Append string s to array a, enlarging as needed static void append(Array& a, const char* s) { if (!s) return; if (!a.size()) a.resize(strlen(s)+1); int len=strlen(&a[0])+strlen(s)+1; if (len>a.size()) { Array tmp(a.size()); strcpy(&tmp[0], &a[0]); a.resize(len*5/4+64); strcpy(&a[0], &tmp[0]); } strcat(&a[0], s); } //////////////////////////// SHA-1 ////////////////////////////// // The SHA1 class is used to compute segment checksums. // SHA-1 code modified from RFC 3174. // http://www.faqs.org/rfcs/rfc3174.html enum { shaSuccess = 0, shaNull, /* Null pointer parameter */ shaInputTooLong, /* input data too long */ shaStateError /* called Input after Result */ }; const int SHA1HashSize=20; class SHA1 { U32 Intermediate_Hash[SHA1HashSize/4]; /* Message Digest */ U32 Length_Low; /* Message length in bits */ U32 Length_High; /* Message length in bits */ int Message_Block_Index; /* Index into message block array */ U8 Message_Block[64]; /* 512-bit message blocks */ int Computed; /* Is the digest computed? */ int Corrupted; /* Is the message digest corrupted? */ U8 result_buf[20]; // Place to put result void SHA1PadMessage(); void SHA1ProcessMessageBlock(); U32 SHA1CircularShift(int bits, U32 word) { return (((word) << (bits)) | ((word) >> (32-(bits)))); } int SHA1Reset(); // Initalize int SHA1Input(const U8 *, unsigned int n); // Hash n bytes int SHA1Result(U8 Message_Digest[SHA1HashSize]); // Store result public: SHA1() {SHA1Reset();} // Begin hash void put(int c) { // Hash 1 byte U8 ch=c; SHA1Input(&ch, 1); } int result(int i); // Finish and return byte i (0..19) of SHA1 hash double size() const { // Number of bytes hashed so far return (Length_Low+4294967296.0*Length_High)/8;} }; int SHA1::result(int i) { assert(i>=0 && i<20); if (!Computed && shaSuccess != SHA1Result(result_buf)) error("SHA1 failed\n"); return result_buf[i]; } /* * SHA1Reset * * Description: * This function will initialize the SHA1Context in preparation * for computing a new SHA1 message digest. * * Parameters: none * * Returns: * sha Error Code. * */ int SHA1::SHA1Reset() { Length_Low = 0; Length_High = 0; Message_Block_Index = 0; Intermediate_Hash[0] = 0x67452301; Intermediate_Hash[1] = 0xEFCDAB89; Intermediate_Hash[2] = 0x98BADCFE; Intermediate_Hash[3] = 0x10325476; Intermediate_Hash[4] = 0xC3D2E1F0; Computed = 0; Corrupted = 0; return shaSuccess; } /* * SHA1Result * * Description: * This function will return the 160-bit message digest into the * Message_Digest array provided by the caller. * NOTE: The first octet of hash is stored in the 0th element, * the last octet of hash in the 19th element. * * Parameters: * Message_Digest: [out] * Where the digest is returned. * * Returns: * sha Error Code. * */ int SHA1::SHA1Result(U8 Message_Digest[SHA1HashSize]) { int i; if (!Message_Digest) { return shaNull; } if (Corrupted) { return Corrupted; } if (!Computed) { SHA1PadMessage(); for(i=0; i<64; ++i) { /* message may be sensitive, clear it out */ Message_Block[i] = 0; } // Length_Low = 0; /* and DON'T clear length */ // Length_High = 0; Computed = 1; } for(i = 0; i < SHA1HashSize; ++i) { Message_Digest[i] = Intermediate_Hash[i>>2] >> 8 * ( 3 - ( i & 0x03 ) ); } return shaSuccess; } /* * SHA1Input * * Description: * This function accepts an array of octets as the next portion * of the message. * * Parameters: * message_array: [in] * An array of characters representing the next portion of * the message. * length: [in] * The length of the message in message_array * * Returns: * sha Error Code. * */ int SHA1::SHA1Input(const U8 *message_array, unsigned length) { if (!length) { return shaSuccess; } if (!message_array) { return shaNull; } if (Computed) { Corrupted = shaStateError; return shaStateError; } if (Corrupted) { return Corrupted; } while(length-- && !Corrupted) { Message_Block[Message_Block_Index++] = (*message_array & 0xFF); Length_Low += 8; if (Length_Low == 0) { Length_High++; if (Length_High == 0) { /* Message is too long */ Corrupted = 1; } } if (Message_Block_Index == 64) { SHA1ProcessMessageBlock(); } message_array++; } return shaSuccess; } /* * SHA1ProcessMessageBlock * * Description: * This function will process the next 512 bits of the message * stored in the Message_Block array. * * Parameters: * None. * * Returns: * Nothing. * * Comments: * Many of the variable names in this code, especially the * single character names, were used because those were the * names used in the publication. * * */ void SHA1::SHA1ProcessMessageBlock() { const U32 K[] = { /* Constants defined in SHA-1 */ 0x5A827999, 0x6ED9EBA1, 0x8F1BBCDC, 0xCA62C1D6 }; int t; /* Loop counter */ U32 temp; /* Temporary word value */ U32 W[80]; /* Word sequence */ U32 A, B, C, D, E; /* Word buffers */ /* * Initialize the first 16 words in the array W */ for(t = 0; t < 16; t++) { W[t] = Message_Block[t * 4] << 24; W[t] |= Message_Block[t * 4 + 1] << 16; W[t] |= Message_Block[t * 4 + 2] << 8; W[t] |= Message_Block[t * 4 + 3]; } for(t = 16; t < 80; t++) { W[t] = SHA1CircularShift(1,W[t-3] ^ W[t-8] ^ W[t-14] ^ W[t-16]); } A = Intermediate_Hash[0]; B = Intermediate_Hash[1]; C = Intermediate_Hash[2]; D = Intermediate_Hash[3]; E = Intermediate_Hash[4]; for(t = 0; t < 20; t++) { temp = SHA1CircularShift(5,A) + ((B & C) | ((~B) & D)) + E + W[t] + K[0]; E = D; D = C; C = SHA1CircularShift(30,B); B = A; A = temp; } for(t = 20; t < 40; t++) { temp = SHA1CircularShift(5,A) + (B ^ C ^ D) + E + W[t] + K[1]; E = D; D = C; C = SHA1CircularShift(30,B); B = A; A = temp; } for(t = 40; t < 60; t++) { temp = SHA1CircularShift(5,A) + ((B & C) | (B & D) | (C & D)) + E + W[t] + K[2]; E = D; D = C; C = SHA1CircularShift(30,B); B = A; A = temp; } for(t = 60; t < 80; t++) { temp = SHA1CircularShift(5,A) + (B ^ C ^ D) + E + W[t] + K[3]; E = D; D = C; C = SHA1CircularShift(30,B); B = A; A = temp; } Intermediate_Hash[0] += A; Intermediate_Hash[1] += B; Intermediate_Hash[2] += C; Intermediate_Hash[3] += D; Intermediate_Hash[4] += E; Message_Block_Index = 0; } /* * SHA1PadMessage * * Description: * According to the standard, the message must be padded to an even * 512 bits. The first padding bit must be a '1'. The last 64 * bits represent the length of the original message. All bits in * between should be 0. This function will pad the message * according to those rules by filling the Message_Block array * accordingly. It will also call the ProcessMessageBlock function * provided appropriately. When it returns, it can be assumed that * the message digest has been computed. * * Parameters: * ProcessMessageBlock: [in] * The appropriate SHA*ProcessMessageBlock function * Returns: * Nothing. * */ void SHA1::SHA1PadMessage() { /* * Check to see if the current message block is too small to hold * the initial padding bits and length. If so, we will pad the * block, process it, and then continue padding into a second * block. */ if (Message_Block_Index > 55) { Message_Block[Message_Block_Index++] = 0x80; while(Message_Block_Index < 64) { Message_Block[Message_Block_Index++] = 0; } SHA1ProcessMessageBlock(); while(Message_Block_Index < 56) { Message_Block[Message_Block_Index++] = 0; } } else { Message_Block[Message_Block_Index++] = 0x80; while(Message_Block_Index < 56) { Message_Block[Message_Block_Index++] = 0; } } /* * Store the message length as the last 8 octets */ Message_Block[56] = Length_High >> 24; Message_Block[57] = Length_High >> 16; Message_Block[58] = Length_High >> 8; Message_Block[59] = Length_High; Message_Block[60] = Length_Low >> 24; Message_Block[61] = Length_Low >> 16; Message_Block[62] = Length_Low >> 8; Message_Block[63] = Length_Low; SHA1ProcessMessageBlock(); } //////////////////////////// ZPAQL ////////////////////////////// // Symbolic constants, instruction size, and names typedef enum {NONE,CONS,CM,ICM,MATCH,AVG,MIX2,MIX,ISSE,SSE} CompType; static const int compsize[256]={0,2,3,2,3,4,6,6,3,5}; // A ZPAQL machine COMP+HCOMP or PCOMP. class ZPAQL { public: ZPAQL(); int read(Reader r); // Read header from archive or array int write(FILE* out); // Write header to archive void inith(); // Initialize as HCOMP to run void initp(); // Initialize as PCOMP to run U32 H(int i) {return h(i);} // get element of h void run(U32 input); // Execute with input FILE* output; // Destination for OUT instruction, or 0 to suppress SHA1* sha1; // Points to checksum computer double memory(); // Return memory requirement in bytes void selectModel(int sel); // Match header to sel // ZPAQ1 block header Array header; // hsize[2] hh hm ph pm n COMP (guard) HCOMP (guard) int cend; // COMP in header[7...cend-1] int hbegin, hend; // HCOMP/PCOMP in header[hbegin...hend-1] int select; // Which optimized version of run()? (default 0) private: // Machine state for executing HCOMP Array m; // memory array M for HCOMP Array h; // hash array H for HCOMP Array r; // 256 element register array U32 a, b, c, d; // machine registers int f; // condition flag int pc; // program counter // Support code void init(int hbits, int mbits); // initialize H and M sizes int execute(); // execute 1 instruction, return 0 after HALT, else 1 void run0(U32 input); // default run() when select==0 void div(U32 x) {if (x) a/=x; else a=0;} void mod(U32 x) {if (x) a%=x; else a=0;} void swap(U32& x) {a^=x; x^=a; a^=x;} void swap(U8& x) {a^=x; x^=a; a^=x;} void err(); // exit with run time error }; // Constructor ZPAQL::ZPAQL() { cend=hbegin=hend=0; // COMP and HCOMP locations a=b=c=d=f=pc=0; // machine state output=0; sha1=0; select=0; } // Read header, return number of bytes read int ZPAQL::read(Reader r) { // Get header size and allocate int hsize=r.get(); hsize+=r.get()*256; header.resize(hsize+300); cend=hbegin=hend=0; header[cend++]=hsize&255; header[cend++]=hsize>>8; while (cend<7) header[cend++]=r.get(); // hh hm ph pm n // Read COMP int n=header[cend-1]; for (int i=0; iheader.size()-8) error("COMP list too big"); for (int j=1; j=7 && cendhbegin && hend=7 && cendhbegin && hend6); init(header[2], header[3]); // hh, hm } // Initialize machine state as PCOMP void ZPAQL::initp() { assert(header.size()>6); init(header[4], header[5]); // ph, pm } // Return memory requirement in bytes double ZPAQL::memory() { double mem=pow(2.0,header[2]+2)+pow(2.0,header[3]) // hh hm +pow(2.0,header[4]+2)+pow(2.0,header[5]) // ph pm +header.size(); int cp=7; // start of comp list for (int i=0; i0); assert(h.size()==0); assert(m.size()==0); assert(cend>=7); assert(hbegin>=cend+128); assert(hend>=hbegin); assert(hend6); assert(hbegin>=cend+128); assert(hend>=hbegin); assert(hend0); assert(h.size()>0); assert(header[0]+256*header[1]==cend+hend-hbegin-2); pc=hbegin; a=input; while (execute()) ; } // Execute one instruction, return 0 after HALT else 1 inline int ZPAQL::execute() { switch(header[pc++]) { case 0: err(); break; // ERROR case 1: ++a; break; // A++ case 2: --a; break; // A-- case 3: a = ~a; break; // A! case 4: a = 0; break; // A=0 case 7: a = r[header[pc++]]; break; // A=R N case 8: swap(b); break; // B<>A case 9: ++b; break; // B++ case 10: --b; break; // B-- case 11: b = ~b; break; // B! case 12: b = 0; break; // B=0 case 15: b = r[header[pc++]]; break; // B=R N case 16: swap(c); break; // C<>A case 17: ++c; break; // C++ case 18: --c; break; // C-- case 19: c = ~c; break; // C! case 20: c = 0; break; // C=0 case 23: c = r[header[pc++]]; break; // C=R N case 24: swap(d); break; // D<>A case 25: ++d; break; // D++ case 26: --d; break; // D-- case 27: d = ~d; break; // D! case 28: d = 0; break; // D=0 case 31: d = r[header[pc++]]; break; // D=R N case 32: swap(m(b)); break; // *B<>A case 33: ++m(b); break; // *B++ case 34: --m(b); break; // *B-- case 35: m(b) = ~m(b); break; // *B! case 36: m(b) = 0; break; // *B=0 case 39: if (f) pc+=((header[pc]+128)&255)-127; else ++pc; break; // JT N case 40: swap(m(c)); break; // *C<>A case 41: ++m(c); break; // *C++ case 42: --m(c); break; // *C-- case 43: m(c) = ~m(c); break; // *C! case 44: m(c) = 0; break; // *C=0 case 47: if (!f) pc+=((header[pc]+128)&255)-127; else ++pc; break; // JF N case 48: swap(h(d)); break; // *D<>A case 49: ++h(d); break; // *D++ case 50: --h(d); break; // *D-- case 51: h(d) = ~h(d); break; // *D! case 52: h(d) = 0; break; // *D=0 case 55: r[header[pc++]] = a; break; // R=A N case 56: return 0 ; // HALT case 57: if (output) putc(a, output); if (sha1) sha1->put(a); break; // OUT case 59: a = (a+m(b)+512)*773; break; // HASH case 60: h(d) = (h(d)+a+512)*773; break; // HASHD case 63: pc+=((header[pc]+128)&255)-127; break; // JMP N case 64: a = a; break; // A=A case 65: a = b; break; // A=B case 66: a = c; break; // A=C case 67: a = d; break; // A=D case 68: a = m(b); break; // A=*B case 69: a = m(c); break; // A=*C case 70: a = h(d); break; // A=*D case 71: a = header[pc++]; break; // A= N case 72: b = a; break; // B=A case 73: b = b; break; // B=B case 74: b = c; break; // B=C case 75: b = d; break; // B=D case 76: b = m(b); break; // B=*B case 77: b = m(c); break; // B=*C case 78: b = h(d); break; // B=*D case 79: b = header[pc++]; break; // B= N case 80: c = a; break; // C=A case 81: c = b; break; // C=B case 82: c = c; break; // C=C case 83: c = d; break; // C=D case 84: c = m(b); break; // C=*B case 85: c = m(c); break; // C=*C case 86: c = h(d); break; // C=*D case 87: c = header[pc++]; break; // C= N case 88: d = a; break; // D=A case 89: d = b; break; // D=B case 90: d = c; break; // D=C case 91: d = d; break; // D=D case 92: d = m(b); break; // D=*B case 93: d = m(c); break; // D=*C case 94: d = h(d); break; // D=*D case 95: d = header[pc++]; break; // D= N case 96: m(b) = a; break; // *B=A case 97: m(b) = b; break; // *B=B case 98: m(b) = c; break; // *B=C case 99: m(b) = d; break; // *B=D case 100: m(b) = m(b); break; // *B=*B case 101: m(b) = m(c); break; // *B=*C case 102: m(b) = h(d); break; // *B=*D case 103: m(b) = header[pc++]; break; // *B= N case 104: m(c) = a; break; // *C=A case 105: m(c) = b; break; // *C=B case 106: m(c) = c; break; // *C=C case 107: m(c) = d; break; // *C=D case 108: m(c) = m(b); break; // *C=*B case 109: m(c) = m(c); break; // *C=*C case 110: m(c) = h(d); break; // *C=*D case 111: m(c) = header[pc++]; break; // *C= N case 112: h(d) = a; break; // *D=A case 113: h(d) = b; break; // *D=B case 114: h(d) = c; break; // *D=C case 115: h(d) = d; break; // *D=D case 116: h(d) = m(b); break; // *D=*B case 117: h(d) = m(c); break; // *D=*C case 118: h(d) = h(d); break; // *D=*D case 119: h(d) = header[pc++]; break; // *D= N case 128: a += a; break; // A+=A case 129: a += b; break; // A+=B case 130: a += c; break; // A+=C case 131: a += d; break; // A+=D case 132: a += m(b); break; // A+=*B case 133: a += m(c); break; // A+=*C case 134: a += h(d); break; // A+=*D case 135: a += header[pc++]; break; // A+= N case 136: a -= a; break; // A-=A case 137: a -= b; break; // A-=B case 138: a -= c; break; // A-=C case 139: a -= d; break; // A-=D case 140: a -= m(b); break; // A-=*B case 141: a -= m(c); break; // A-=*C case 142: a -= h(d); break; // A-=*D case 143: a -= header[pc++]; break; // A-= N case 144: a *= a; break; // A*=A case 145: a *= b; break; // A*=B case 146: a *= c; break; // A*=C case 147: a *= d; break; // A*=D case 148: a *= m(b); break; // A*=*B case 149: a *= m(c); break; // A*=*C case 150: a *= h(d); break; // A*=*D case 151: a *= header[pc++]; break; // A*= N case 152: div(a); break; // A/=A case 153: div(b); break; // A/=B case 154: div(c); break; // A/=C case 155: div(d); break; // A/=D case 156: div(m(b)); break; // A/=*B case 157: div(m(c)); break; // A/=*C case 158: div(h(d)); break; // A/=*D case 159: div(header[pc++]); break; // A/= N case 160: mod(a); break; // A%=A case 161: mod(b); break; // A%=B case 162: mod(c); break; // A%=C case 163: mod(d); break; // A%=D case 164: mod(m(b)); break; // A%=*B case 165: mod(m(c)); break; // A%=*C case 166: mod(h(d)); break; // A%=*D case 167: mod(header[pc++]); break; // A%= N case 168: a &= a; break; // A&=A case 169: a &= b; break; // A&=B case 170: a &= c; break; // A&=C case 171: a &= d; break; // A&=D case 172: a &= m(b); break; // A&=*B case 173: a &= m(c); break; // A&=*C case 174: a &= h(d); break; // A&=*D case 175: a &= header[pc++]; break; // A&= N case 176: a &= ~ a; break; // A&~A case 177: a &= ~ b; break; // A&~B case 178: a &= ~ c; break; // A&~C case 179: a &= ~ d; break; // A&~D case 180: a &= ~ m(b); break; // A&~*B case 181: a &= ~ m(c); break; // A&~*C case 182: a &= ~ h(d); break; // A&~*D case 183: a &= ~ header[pc++]; break; // A&~ N case 184: a |= a; break; // A|=A case 185: a |= b; break; // A|=B case 186: a |= c; break; // A|=C case 187: a |= d; break; // A|=D case 188: a |= m(b); break; // A|=*B case 189: a |= m(c); break; // A|=*C case 190: a |= h(d); break; // A|=*D case 191: a |= header[pc++]; break; // A|= N case 192: a ^= a; break; // A^=A case 193: a ^= b; break; // A^=B case 194: a ^= c; break; // A^=C case 195: a ^= d; break; // A^=D case 196: a ^= m(b); break; // A^=*B case 197: a ^= m(c); break; // A^=*C case 198: a ^= h(d); break; // A^=*D case 199: a ^= header[pc++]; break; // A^= N case 200: a <<= (a&31); break; // A<<=A case 201: a <<= (b&31); break; // A<<=B case 202: a <<= (c&31); break; // A<<=C case 203: a <<= (d&31); break; // A<<=D case 204: a <<= (m(b)&31); break; // A<<=*B case 205: a <<= (m(c)&31); break; // A<<=*C case 206: a <<= (h(d)&31); break; // A<<=*D case 207: a <<= (header[pc++]&31); break; // A<<= N case 208: a >>= (a&31); break; // A>>=A case 209: a >>= (b&31); break; // A>>=B case 210: a >>= (c&31); break; // A>>=C case 211: a >>= (d&31); break; // A>>=D case 212: a >>= (m(b)&31); break; // A>>=*B case 213: a >>= (m(c)&31); break; // A>>=*C case 214: a >>= (h(d)&31); break; // A>>=*D case 215: a >>= (header[pc++]&31); break; // A>>= N case 216: f = (a == a); break; // A==A case 217: f = (a == b); break; // A==B case 218: f = (a == c); break; // A==C case 219: f = (a == d); break; // A==D case 220: f = (a == U32(m(b))); break; // A==*B case 221: f = (a == U32(m(c))); break; // A==*C case 222: f = (a == h(d)); break; // A==*D case 223: f = (a == U32(header[pc++])); break; // A== N case 224: f = (a < a); break; // A a); break; // A>A case 233: f = (a > b); break; // A>B case 234: f = (a > c); break; // A>C case 235: f = (a > d); break; // A>D case 236: f = (a > U32(m(b))); break; // A>*B case 237: f = (a > U32(m(c))); break; // A>*C case 238: f = (a > h(d)); break; // A>*D case 239: f = (a > U32(header[pc++])); break; // A> N case 255: if((pc=hbegin+header[pc]+256*header[pc+1])>=hend)err();break;//LJ default: err(); } return 1; } // Print illegal instruction error message and exit void ZPAQL::err() { error("ZPAQL execution error"); } //////////////////////////// Component //////////////////////////// // A Component is a context model, indirect context model, match model, // fixed weight mixer, adaptive 2 input mixer without or with current // partial byte as context, adaptive m input mixer (without or with), // or SSE (without or with). struct Component { int limit; // max count for cm U32 cxt; // saved context int a, b, c; // multi-purpose variables Array cm; // cm[cxt] -> p in bits 31..10, n in 9..0; MATCH index Array ht; // ICM hash table[0..size1][0..15] of bit histories; MATCH buf Array a16; // multi-use Component(); // initialize to all 0 }; Component::Component(): limit(0), cxt(0), a(0), b(0), c(0) {} ////////////////////////// StateTable ////////////////////////// // Next state table generator class StateTable { enum {B=6, N=64}; // sizes of b, t static U8 ns[1024]; // state*4 -> next state if 0, if 1, n0, n1 static const int bound[B]; // n0 -> max n1, n1 -> max n0 int num_states(int n0, int n1); // compute t[n0][n1][1] void discount(int& n0); // set new value of n0 after 1 or n1 after 0 void next_state(int& n0, int& n1, int y); // new (n0,n1) after bit y public: int next(int state, int y) { // next state for bit y assert(state>=0 && state<256); assert(y>=0 && y<4); return ns[state*4+y]; } int cminit(int state) { // initial probability of 1 * 2^23 assert(state>=0 && state<256); return ((ns[state*4+3]*2+1)<<22)/(ns[state*4+2]+ns[state*4+3]+1); } StateTable(); }; U8 StateTable::ns[1024]={0}; const int StateTable::bound[B]={20,48,15,8,6,5}; // n0 -> max n1, n1 -> max n0 // How many states with count of n0 zeros, n1 ones (0...2) int StateTable::num_states(int n0, int n1) { if (n0=N || n1>=N || n1>=B || n0>bound[n1]) return 0; return 1+(n1>0 && n0+n1<=17); } // New value of count n0 if 1 is observed (and vice versa) void StateTable::discount(int& n0) { n0=(n0>=1)+(n0>=2)+(n0>=3)+(n0>=4)+(n0>=5)+(n0>=7)+(n0>=8); } // compute next n0,n1 (0 to N) given input y (0 or 1) void StateTable::next_state(int& n0, int& n1, int y) { if (n0 20,0 // 48,1,0 -> 48,1 // 15,2,0 -> 8,1 // 8,3,0 -> 6,2 // 8,3,1 -> 5,3 // 6,4,0 -> 5,3 // 5,5,0 -> 5,4 // 5,5,1 -> 4,5 while (!num_states(n0, n1)) { if (n1<2) --n0; else { n0=(n0*(n1-1)+(n1/2))/n1; --n1; } } } } // Initialize next state table ns[state*4] -> next if 0, next if 1, n0, n1 StateTable::StateTable() { // Assign states by increasing priority U8 t[N][N][2]={{{0}}}; // (n0,n1,y) -> state number int state=0; for (int i=0; i=0 && n<=2); if (n) { t[n0][n1][0]=state; t[n0][n1][1]=state+n-1; state+=n; } } } // Generate next state table for (int n0=0; n0=0 && s<256); int s0=n0, s1=n1; next_state(s0, s1, 0); assert(s0>=0 && s0=0 && s1=0 && s0=0 && s1>17); pn+=(error*dt[count]&-1024)+(count floor(32768/(1+exp(-x/64))) int squash(int x) { assert(x>=-2048 && x<=2047); return squasht[x+2048]; } // x -> round(64*log((x+0.5)/(32767.5-x))), approx inverse of squash int stretch(int x) { assert(x>=0 && x<=32767); return stretcht[x]; } // bound x to a 12 bit signed int int clamp2k(int x) { if (x<-2048) return -2048; else if (x>2047) return 2047; else return x; } // bound x to a 20 bit signed int int clamp512k(int x) { if (x<-(1<<19)) return -(1<<19); else if (x>=(1<<19)) return (1<<19)-1; else return x; } // Get cxt in ht, creating a new row if needed int find(Array& ht, int sizebits, U32 cxt); }; // Initailize the model Predictor::Predictor(ZPAQL& zr): c8(1), hmap4(1), z(zr) { assert(sizeof(U8)==1); assert(sizeof(U16)==2); assert(sizeof(U32)==4); assert(sizeof(short)==2); assert(sizeof(int)==4); // Initialize tables for (int i=0; i<1024; ++i) dt[i]=(1<<17)/(i*2+3)*2; for (int i=0; i<32768; ++i) stretcht[i]=int(log((i+0.5)/(32767.5-i))*64+0.5+100000)-100000; for (int i=0; i<4096; ++i) squasht[i]=int(32768.0/(1+exp((i-2048)*(-1.0/64)))); // Verify floating point math for squash() and stretch() U32 sqsum=0, stsum=0; for (int i=32767; i>=0; --i) stsum=stsum*3+stretch(i); for (int i=4095; i>=0; --i) sqsum=sqsum*3+squash(i-2048); assert(stsum==3887533746u); assert(sqsum==2278286169u); // Initialize context hash function z.inith(); // Initialize predictions for (int i=0; i<256; ++i) p[i]=0; // Initialize components int n=z.header[6]; // hsize[0..1] hh hm ph pm n (comp)[n] END 0[128] (hcomp) END if (n<1 || n>255) error("n must be 1..255 components"); const U8* cp=&z.header[7]; // start of component list for (int i=0; i&z.header[0] && cp<&z.header[z.header.size()-8]); Component& cr=comp[i]; switch(cp[0]) { case CONS: // c p[i]=(cp[1]-128)*4; break; case CM: // sizebits limit cr.cm.resize(1, cp[1]); // packed CM (22 bits) + CMCOUNT (10 bits) cr.limit=cp[2]*4; for (int j=0; j=i) error("MIX2 k >= i"); if (cp[2]>=i) error("MIX2 j >= i"); cr.c=(1<=i) error("MIX j >= i"); if (cp[3]<1 || cp[3]>i-cp[2]) error("MIX m not in 1..i-j"); int m=cp[3]; // number of inputs assert(m>=1); cr.c=(1<=i) error("ISSE j >= i"); cr.ht.resize(64, cp[1]); cr.cm.resize(512); for (int j=0; j<256; ++j) { cr.cm[j*2]=1<<15; cr.cm[j*2+1]=clamp512k(stretch(st.cminit(j)>>8)<<10); } break; case SSE: // sizebits j start limit if (cp[2]>=i) error("SSE j >= i"); if (cp[3]>cp[4]*4) error("SSE start > limit*4"); cr.cm.resize(32, cp[1]); cr.limit=cp[4]*4; for (int j=0; j0); cp+=compsize[*cp]; assert(cp>=&z.header[7] && cp<&z.header[z.cend]); } } int Predictor::predict0() { assert(c8>=1 && c8<=255); // Predict next bit int n=z.header[6]; assert(n>0 && n<=255); const U8* cp=&z.header[7]; assert(cp[-1]==n); for (int i=0; i&z.header[0] && cp<&z.header[z.header.size()-8]); Component& cr=comp[i]; switch(cp[0]) { case CONS: // c break; case CM: // sizebits limit cr.cxt=z.H(i)^hmap4; p[i]=stretch(cr.cm(cr.cxt)>>17); break; case ICM: // sizebits assert((hmap4&15)>0); if (c8==1 || (c8&0xf0)==16) cr.c=find(cr.ht, cp[1]+2, z.H(i)+16*c8); cr.cxt=cr.ht[cr.c+(hmap4&15)]; p[i]=stretch(cr.cm(cr.cxt)>>8); break; case MATCH: // sizebits bufbits: a=len, b=offset, c=bit, cxt=256/len, // ht=buf, limit=8*pos+bp assert(cr.a>=0 && cr.a<=255); if (cr.a==0) p[i]=0; else { cr.c=cr.ht((cr.limit>>3)-cr.b)>>(7-(cr.limit&7))&1; // predicted bit p[i]=stretch(cr.cxt*(cr.c*-2+1)&32767); } break; case AVG: // j k wt p[i]=(p[cp[1]]*cp[3]+p[cp[2]]*(256-cp[3]))>>8; break; case MIX2: { // sizebits j k rate mask // c=size cm=wt[size][m] cxt=input cr.cxt=((z.H(i)+(c8&cp[5]))&(cr.c-1)); assert(int(cr.cxt)>=0 && int(cr.cxt)=0 && w<65536); p[i]=(w*p[cp[2]]+(65536-w)*p[cp[3]])>>16; assert(p[i]>=-2048 && p[i]<2048); } break; case MIX: { // sizebits j m rate mask // c=size cm=wt[size][m] cxt=index of wt in cm int m=cp[3]; assert(m>=1 && m<=i); cr.cxt=z.H(i)+(c8&cp[5]); cr.cxt=(cr.cxt&(cr.c-1))*m; // pointer to row of weights assert(int(cr.cxt)>=0 && int(cr.cxt)<=cr.cm.size()-m); int* wt=(int*)&cr.cm[cr.cxt]; p[i]=0; for (int j=0; j>8)*p[cp[2]+j]; p[i]=clamp2k(p[i]>>8); } break; case ISSE: { // sizebits j -- c=hi, cxt=bh assert((hmap4&15)>0); if (c8==1 || (c8&0xf0)==16) cr.c=find(cr.ht, cp[1]+2, z.H(i)+16*c8); cr.cxt=cr.ht[cr.c+(hmap4&15)]; // bit history int *wt=(int*)&cr.cm[cr.cxt*2]; p[i]=clamp2k((wt[0]*p[cp[2]]+wt[1]*64)>>16); } break; case SSE: { // sizebits j start limit cr.cxt=(z.H(i)+c8)*32; int pq=p[cp[2]]+992; if (pq<0) pq=0; if (pq>1983) pq=1983; int wt=pq&63; pq>>=6; assert(pq>=0 && pq<=30); cr.cxt+=pq; p[i]=stretch(((cr.cm(cr.cxt)>>10)*(64-wt)+(cr.cm(cr.cxt+1)>>10)*wt)>>13); cr.cxt+=wt>>5; } break; default: error("component predict not implemented"); } cp+=compsize[cp[0]]; assert(cp<&z.header[z.cend]); assert(p[i]>=-2048 && p[i]<2048); } assert(cp[0]==NONE); return squash(p[n-1]); } // Update model with decoded bit y (0...1) void Predictor::update0(int y) { assert(y==0 || y==1); assert(c8>=1 && c8<=255); assert(hmap4>=1 && hmap4<=511); // Update components const U8* cp=&z.header[7]; int n=z.header[6]; assert(n>=1 && n<=255); assert(cp[-1]==n); for (int i=0; i>8))>>2; } break; case MATCH: // sizebits bufbits: // a=len, b=offset, c=bit, cm=index, cxt=256/len // ht=buf, limit=8*pos+bp { assert(cr.a>=0 && cr.a<=255); assert(cr.c==0 || cr.c==1); if (cr.c!=y) cr.a=0; // mismatch? cr.ht(cr.limit>>3)+=cr.ht(cr.limit>>3)+y; if ((++cr.limit&7)==0) { int pos=cr.limit>>3; if (cr.a==0) { // look for a match cr.b=pos-cr.cm(z.H(i)); if (cr.b&(cr.ht.size()-1)) while (cr.a<255 && cr.ht(pos-cr.a-1)==cr.ht(pos-cr.a-cr.b-1)) ++cr.a; } else cr.a+=cr.a<255; cr.cm(z.H(i))=pos; if (cr.a>0) cr.cxt=2048/cr.a; } } break; case AVG: // j k wt break; case MIX2: { // sizebits j k rate mask // cm=input[2],wt[size][2], cxt=weight row assert(cr.a16.size()==cr.c); assert(int(cr.cxt)>=0 && int(cr.cxt)>5; int w=cr.a16[cr.cxt]; w+=(err*(p[cp[2]]-p[cp[3]])+(1<<12))>>13; if (w<0) w=0; if (w>65535) w=65535; cr.a16[cr.cxt]=w; } break; case MIX: { // sizebits j m rate mask // cm=wt[size][m], cxt=input int m=cp[3]; assert(m>0 && m<=i); assert(cr.cm.size()==m*cr.c); assert(int(cr.cxt)>=0 && int(cr.cxt)<=cr.cm.size()-m); int err=(y*32767-squash(p[i]))*cp[4]>>4; int* wt=(int*)&cr.cm[cr.cxt]; for (int j=0; j>13)); } break; case ISSE: { // sizebits j -- c=hi, cxt=bh assert(cr.cxt==cr.ht[cr.c+(hmap4&15)]); int err=y*32767-squash(p[i]); int *wt=(int*)&cr.cm[cr.cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[cp[2]]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); cr.ht[cr.c+(hmap4&15)]=st.next(cr.cxt, y); } break; case SSE: // sizebits j start limit train(cr, y); break; default: assert(0); } cp+=compsize[cp[0]]; assert(cp>=&z.header[7] && cp<&z.header[z.cend] && cp<&z.header[z.header.size()-8]); } assert(cp[0]==NONE); // Save bit y in c8, hmap4 c8+=c8+y; if (c8>=256) { z.run(c8-256); hmap4=1; c8=1; } else if (c8>=16 && c8<32) hmap4=(hmap4&0xf)<<5|y<<4|1; else hmap4=(hmap4&0x1f0)|(((hmap4&0xf)*2+y)&0xf); } // Find cxt row in hash table ht. ht has rows of 16 indexed by the // low sizebits of cxt with element 0 having the next higher 8 bits for // collision detection. If not found after 3 adjacent tries, replace the // row with lowest element 1 as priority. Return index of row. int Predictor::find(Array& ht, int sizebits, U32 cxt) { assert(ht.size()==16<>sizebits&255; int h0=(cxt*16)&(ht.size()-16); if (ht[h0]==chk) return h0; int h1=h0^16; if (ht[h1]==chk) return h1; int h2=h0^32; if (ht[h2]==chk) return h2; if (ht[h0+1]<=ht[h1+1] && ht[h0+1]<=ht[h2+1]) return memset(&ht[h0], 0, 16), ht[h0]=chk, h0; else if (ht[h1+1] 0 then load the selected header and set select=sel. // Otherwise search header for an optimization and set select>0 if found. void ZPAQL::selectModel(int sel) { // A list of headers for which optimizations are available static const U8 models[]={ // fast.cfg 26,0,1,2,0,0,2,3,16,8,19,0,0, // HCOMP 96,4,28, 59,10,59,112,25,10,59,10,59,112,56,0, // mid.cfg 69,0,3,3,0,0,8,3,5,8,13,0,8,17,1,8, 18,2,8,18,3,8,19,4,4,22,24,7,16,0,7,24, 255,0, // HCOMP 17,104,74,4,95,1,59,112,10,25,59,112,10,25,59,112, 10,25,59,112,10,25,59,112,10,25,59,10,59,112,25,69, 207,8,112,56,0, // max.cfg 196,0,5,9,0,0,22,1,160,3,5,8,13,1,8,16, 2,8,18,3,8,19,4,8,19,5,8,20,6,4,22,24, 3,17,8,19,9,3,13,3,13,3,13,3,14,7,16,0, 15,24,255,7,8,0,16,10,255,6,0,15,16,24,0,9, 8,17,32,255,6,8,17,18,16,255,9,16,19,32,255,6, 0,19,20,16,0,0, // HCOMP 17,104,74,4,95,2,59,112,10,25, 59,112,10,25,59,112,10,25,59,112,10,25,59,112,10,25, 59,10,59,112,10,25,59,112,10,25,69,183,32,239,64,47, 14,231,91,47,10,25,60,26,48,134,151,20,112,63,9,70, 223,0,39,3,25,112,26,52,25,25,74,10,4,59,112,25, 10,4,59,112,25,10,4,59,112,25,65,143,212,72,4,59, 112,8,143,216,8,68,175,60,60,25,69,207,9,112,25,25, 25,25,25,112,56,0, // end of list 0,0}; // If sel>0 then load the selected optimized header int p=0, len=0, count=0; while (p<=int(sizeof(models))-2) { ++count; len=models[p]+256*models[p+1]; if (len<1) break; if (sel>0 && count==sel) { // load header read(Reader(models+p, len+2)); select=count; break; } else if (sel==0) { if (cend+hend-hbegin==len+2 && memcmp(&header[0], models+p, cend)==0 && memcmp(&header[hbegin], models+p+cend, hend-hbegin)==0) { select=count; } } p+=len+2; } if (cend<7) error("Invalid compression option"); } // Optimized predict int Predictor::predict() { switch(z.select) { // fast.cfg case 1: { // 2 components // 0 ICM 16 if (c8==1 || (c8&0xf0)==16) comp[0].c=find(comp[0].ht, 16+2, z.H(0)+16*c8); comp[0].cxt=comp[0].ht[comp[0].c+(hmap4&15)]; p[0]=stretch(comp[0].cm(comp[0].cxt)>>8); // 1 ISSE 19 0 { if (c8==1 || (c8&0xf0)==16) comp[1].c=find(comp[1].ht, 21, z.H(1)+16*c8); comp[1].cxt=comp[1].ht[comp[1].c+(hmap4&15)]; int *wt=(int*)&comp[1].cm[comp[1].cxt*2]; p[1]=clamp2k((wt[0]*p[0]+wt[1]*64)>>16); } return squash(p[1]); } // mid.cfg case 2: { // 8 components // 0 ICM 5 if (c8==1 || (c8&0xf0)==16) comp[0].c=find(comp[0].ht, 5+2, z.H(0)+16*c8); comp[0].cxt=comp[0].ht[comp[0].c+(hmap4&15)]; p[0]=stretch(comp[0].cm(comp[0].cxt)>>8); // 1 ISSE 13 0 { if (c8==1 || (c8&0xf0)==16) comp[1].c=find(comp[1].ht, 15, z.H(1)+16*c8); comp[1].cxt=comp[1].ht[comp[1].c+(hmap4&15)]; int *wt=(int*)&comp[1].cm[comp[1].cxt*2]; p[1]=clamp2k((wt[0]*p[0]+wt[1]*64)>>16); } // 2 ISSE 17 1 { if (c8==1 || (c8&0xf0)==16) comp[2].c=find(comp[2].ht, 19, z.H(2)+16*c8); comp[2].cxt=comp[2].ht[comp[2].c+(hmap4&15)]; int *wt=(int*)&comp[2].cm[comp[2].cxt*2]; p[2]=clamp2k((wt[0]*p[1]+wt[1]*64)>>16); } // 3 ISSE 18 2 { if (c8==1 || (c8&0xf0)==16) comp[3].c=find(comp[3].ht, 20, z.H(3)+16*c8); comp[3].cxt=comp[3].ht[comp[3].c+(hmap4&15)]; int *wt=(int*)&comp[3].cm[comp[3].cxt*2]; p[3]=clamp2k((wt[0]*p[2]+wt[1]*64)>>16); } // 4 ISSE 18 3 { if (c8==1 || (c8&0xf0)==16) comp[4].c=find(comp[4].ht, 20, z.H(4)+16*c8); comp[4].cxt=comp[4].ht[comp[4].c+(hmap4&15)]; int *wt=(int*)&comp[4].cm[comp[4].cxt*2]; p[4]=clamp2k((wt[0]*p[3]+wt[1]*64)>>16); } // 5 ISSE 19 4 { if (c8==1 || (c8&0xf0)==16) comp[5].c=find(comp[5].ht, 21, z.H(5)+16*c8); comp[5].cxt=comp[5].ht[comp[5].c+(hmap4&15)]; int *wt=(int*)&comp[5].cm[comp[5].cxt*2]; p[5]=clamp2k((wt[0]*p[4]+wt[1]*64)>>16); } // 6 MATCH 22 24 if (comp[6].a==0) p[6]=0; else { comp[6].c=comp[6].ht((comp[6].limit>>3) -comp[6].b)>>(7-(comp[6].limit&7))&1; p[6]=stretch(comp[6].cxt*(comp[6].c*-2+1)&32767); } // 7 MIX 16 0 7 24 255 { comp[7].cxt=z.H(7)+(c8&255); comp[7].cxt=(comp[7].cxt&(comp[7].c-1))*7; int* wt=(int*)&comp[7].cm[comp[7].cxt]; p[7]=(wt[0]>>8)*p[0]; p[7]+=(wt[1]>>8)*p[1]; p[7]+=(wt[2]>>8)*p[2]; p[7]+=(wt[3]>>8)*p[3]; p[7]+=(wt[4]>>8)*p[4]; p[7]+=(wt[5]>>8)*p[5]; p[7]+=(wt[6]>>8)*p[6]; p[7]=clamp2k(p[7]>>8); } return squash(p[7]); } // max.cfg case 3: { // 22 components // 0 CONST 160 // 1 ICM 5 if (c8==1 || (c8&0xf0)==16) comp[1].c=find(comp[1].ht, 5+2, z.H(1)+16*c8); comp[1].cxt=comp[1].ht[comp[1].c+(hmap4&15)]; p[1]=stretch(comp[1].cm(comp[1].cxt)>>8); // 2 ISSE 13 1 { if (c8==1 || (c8&0xf0)==16) comp[2].c=find(comp[2].ht, 15, z.H(2)+16*c8); comp[2].cxt=comp[2].ht[comp[2].c+(hmap4&15)]; int *wt=(int*)&comp[2].cm[comp[2].cxt*2]; p[2]=clamp2k((wt[0]*p[1]+wt[1]*64)>>16); } // 3 ISSE 16 2 { if (c8==1 || (c8&0xf0)==16) comp[3].c=find(comp[3].ht, 18, z.H(3)+16*c8); comp[3].cxt=comp[3].ht[comp[3].c+(hmap4&15)]; int *wt=(int*)&comp[3].cm[comp[3].cxt*2]; p[3]=clamp2k((wt[0]*p[2]+wt[1]*64)>>16); } // 4 ISSE 18 3 { if (c8==1 || (c8&0xf0)==16) comp[4].c=find(comp[4].ht, 20, z.H(4)+16*c8); comp[4].cxt=comp[4].ht[comp[4].c+(hmap4&15)]; int *wt=(int*)&comp[4].cm[comp[4].cxt*2]; p[4]=clamp2k((wt[0]*p[3]+wt[1]*64)>>16); } // 5 ISSE 19 4 { if (c8==1 || (c8&0xf0)==16) comp[5].c=find(comp[5].ht, 21, z.H(5)+16*c8); comp[5].cxt=comp[5].ht[comp[5].c+(hmap4&15)]; int *wt=(int*)&comp[5].cm[comp[5].cxt*2]; p[5]=clamp2k((wt[0]*p[4]+wt[1]*64)>>16); } // 6 ISSE 19 5 { if (c8==1 || (c8&0xf0)==16) comp[6].c=find(comp[6].ht, 21, z.H(6)+16*c8); comp[6].cxt=comp[6].ht[comp[6].c+(hmap4&15)]; int *wt=(int*)&comp[6].cm[comp[6].cxt*2]; p[6]=clamp2k((wt[0]*p[5]+wt[1]*64)>>16); } // 7 ISSE 20 6 { if (c8==1 || (c8&0xf0)==16) comp[7].c=find(comp[7].ht, 22, z.H(7)+16*c8); comp[7].cxt=comp[7].ht[comp[7].c+(hmap4&15)]; int *wt=(int*)&comp[7].cm[comp[7].cxt*2]; p[7]=clamp2k((wt[0]*p[6]+wt[1]*64)>>16); } // 8 MATCH 22 24 if (comp[8].a==0) p[8]=0; else { comp[8].c=comp[8].ht((comp[8].limit>>3) -comp[8].b)>>(7-(comp[8].limit&7))&1; p[8]=stretch(comp[8].cxt*(comp[8].c*-2+1)&32767); } // 9 ICM 17 if (c8==1 || (c8&0xf0)==16) comp[9].c=find(comp[9].ht, 17+2, z.H(9)+16*c8); comp[9].cxt=comp[9].ht[comp[9].c+(hmap4&15)]; p[9]=stretch(comp[9].cm(comp[9].cxt)>>8); // 10 ISSE 19 9 { if (c8==1 || (c8&0xf0)==16) comp[10].c=find(comp[10].ht, 21, z.H(10)+16*c8); comp[10].cxt=comp[10].ht[comp[10].c+(hmap4&15)]; int *wt=(int*)&comp[10].cm[comp[10].cxt*2]; p[10]=clamp2k((wt[0]*p[9]+wt[1]*64)>>16); } // 11 ICM 13 if (c8==1 || (c8&0xf0)==16) comp[11].c=find(comp[11].ht, 13+2, z.H(11)+16*c8); comp[11].cxt=comp[11].ht[comp[11].c+(hmap4&15)]; p[11]=stretch(comp[11].cm(comp[11].cxt)>>8); // 12 ICM 13 if (c8==1 || (c8&0xf0)==16) comp[12].c=find(comp[12].ht, 13+2, z.H(12)+16*c8); comp[12].cxt=comp[12].ht[comp[12].c+(hmap4&15)]; p[12]=stretch(comp[12].cm(comp[12].cxt)>>8); // 13 ICM 13 if (c8==1 || (c8&0xf0)==16) comp[13].c=find(comp[13].ht, 13+2, z.H(13)+16*c8); comp[13].cxt=comp[13].ht[comp[13].c+(hmap4&15)]; p[13]=stretch(comp[13].cm(comp[13].cxt)>>8); // 14 ICM 14 if (c8==1 || (c8&0xf0)==16) comp[14].c=find(comp[14].ht, 14+2, z.H(14)+16*c8); comp[14].cxt=comp[14].ht[comp[14].c+(hmap4&15)]; p[14]=stretch(comp[14].cm(comp[14].cxt)>>8); // 15 MIX 16 0 15 24 255 { comp[15].cxt=z.H(15)+(c8&255); comp[15].cxt=(comp[15].cxt&(comp[15].c-1))*15; int* wt=(int*)&comp[15].cm[comp[15].cxt]; p[15]=(wt[0]>>8)*p[0]; p[15]+=(wt[1]>>8)*p[1]; p[15]+=(wt[2]>>8)*p[2]; p[15]+=(wt[3]>>8)*p[3]; p[15]+=(wt[4]>>8)*p[4]; p[15]+=(wt[5]>>8)*p[5]; p[15]+=(wt[6]>>8)*p[6]; p[15]+=(wt[7]>>8)*p[7]; p[15]+=(wt[8]>>8)*p[8]; p[15]+=(wt[9]>>8)*p[9]; p[15]+=(wt[10]>>8)*p[10]; p[15]+=(wt[11]>>8)*p[11]; p[15]+=(wt[12]>>8)*p[12]; p[15]+=(wt[13]>>8)*p[13]; p[15]+=(wt[14]>>8)*p[14]; p[15]=clamp2k(p[15]>>8); } // 16 MIX 8 0 16 10 255 { comp[16].cxt=z.H(16)+(c8&255); comp[16].cxt=(comp[16].cxt&(comp[16].c-1))*16; int* wt=(int*)&comp[16].cm[comp[16].cxt]; p[16]=(wt[0]>>8)*p[0]; p[16]+=(wt[1]>>8)*p[1]; p[16]+=(wt[2]>>8)*p[2]; p[16]+=(wt[3]>>8)*p[3]; p[16]+=(wt[4]>>8)*p[4]; p[16]+=(wt[5]>>8)*p[5]; p[16]+=(wt[6]>>8)*p[6]; p[16]+=(wt[7]>>8)*p[7]; p[16]+=(wt[8]>>8)*p[8]; p[16]+=(wt[9]>>8)*p[9]; p[16]+=(wt[10]>>8)*p[10]; p[16]+=(wt[11]>>8)*p[11]; p[16]+=(wt[12]>>8)*p[12]; p[16]+=(wt[13]>>8)*p[13]; p[16]+=(wt[14]>>8)*p[14]; p[16]+=(wt[15]>>8)*p[15]; p[16]=clamp2k(p[16]>>8); } // 17 MIX2 0 15 16 24 0 { comp[17].cxt=((z.H(17)+(c8&0))&(comp[17].c-1)); int w=comp[17].a16[comp[17].cxt]; p[17]=(w*p[15]+(65536-w)*p[16])>>16; } // 18 SSE 8 17 32 255 { comp[18].cxt=(z.H(18)+c8)*32; int pq=p[17]+992; if (pq<0) pq=0; if (pq>1983) pq=1983; int wt=pq&63; pq>>=6; comp[18].cxt+=pq; p[18]=stretch(((comp[18].cm(comp[18].cxt)>>10)*(64-wt) +(comp[18].cm(comp[18].cxt+1)>>10)*wt)>>13); comp[18].cxt+=wt>>5; } // 19 MIX2 8 17 18 16 255 { comp[19].cxt=((z.H(19)+(c8&255))&(comp[19].c-1)); int w=comp[19].a16[comp[19].cxt]; p[19]=(w*p[17]+(65536-w)*p[18])>>16; } // 20 SSE 16 19 32 255 { comp[20].cxt=(z.H(20)+c8)*32; int pq=p[19]+992; if (pq<0) pq=0; if (pq>1983) pq=1983; int wt=pq&63; pq>>=6; comp[20].cxt+=pq; p[20]=stretch(((comp[20].cm(comp[20].cxt)>>10)*(64-wt) +(comp[20].cm(comp[20].cxt+1)>>10)*wt)>>13); comp[20].cxt+=wt>>5; } // 21 MIX2 0 19 20 16 0 { comp[21].cxt=((z.H(21)+(c8&0))&(comp[21].c-1)); int w=comp[21].a16[comp[21].cxt]; p[21]=(w*p[19]+(65536-w)*p[20])>>16; } return squash(p[21]); } // Not optimized default: return predict0(); } } void Predictor::update(int y) { switch(z.select) { // fast.cfg case 1: { // 2 components // 0 ICM 16 { comp[0].ht[comp[0].c+(hmap4&15)]= st.next(comp[0].ht[comp[0].c+(hmap4&15)], y); U32& pn=comp[0].cm(comp[0].cxt); pn+=int(y*32767-(pn>>8))>>2; } // 1 ISSE 19 0 { int err=y*32767-squash(p[1]); int *wt=(int*)&comp[1].cm[comp[1].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[0]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[1].ht[comp[1].c+(hmap4&15)]=st.next(comp[1].cxt, y); } break; } // mid.cfg case 2: { // 8 components // 0 ICM 5 { comp[0].ht[comp[0].c+(hmap4&15)]= st.next(comp[0].ht[comp[0].c+(hmap4&15)], y); U32& pn=comp[0].cm(comp[0].cxt); pn+=int(y*32767-(pn>>8))>>2; } // 1 ISSE 13 0 { int err=y*32767-squash(p[1]); int *wt=(int*)&comp[1].cm[comp[1].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[0]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[1].ht[comp[1].c+(hmap4&15)]=st.next(comp[1].cxt, y); } // 2 ISSE 17 1 { int err=y*32767-squash(p[2]); int *wt=(int*)&comp[2].cm[comp[2].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[1]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[2].ht[comp[2].c+(hmap4&15)]=st.next(comp[2].cxt, y); } // 3 ISSE 18 2 { int err=y*32767-squash(p[3]); int *wt=(int*)&comp[3].cm[comp[3].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[2]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[3].ht[comp[3].c+(hmap4&15)]=st.next(comp[3].cxt, y); } // 4 ISSE 18 3 { int err=y*32767-squash(p[4]); int *wt=(int*)&comp[4].cm[comp[4].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[3]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[4].ht[comp[4].c+(hmap4&15)]=st.next(comp[4].cxt, y); } // 5 ISSE 19 4 { int err=y*32767-squash(p[5]); int *wt=(int*)&comp[5].cm[comp[5].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[4]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[5].ht[comp[5].c+(hmap4&15)]=st.next(comp[5].cxt, y); } // 6 MATCH 22 24 { if (comp[6].c!=y) comp[6].a=0; comp[6].ht(comp[6].limit>>3)+=comp[6].ht(comp[6].limit>>3)+y; if ((++comp[6].limit&7)==0) { int pos=comp[6].limit>>3; if (comp[6].a==0) { comp[6].b=pos-comp[6].cm(z.H(6)); if (comp[6].b&(comp[6].ht.size()-1)) while (comp[6].a<255 && comp[6].ht(pos-comp[6].a-1) ==comp[6].ht(pos-comp[6].a-comp[6].b-1)) ++comp[6].a; } else comp[6].a+=comp[6].a<255; comp[6].cm(z.H(6))=pos; if (comp[6].a>0) comp[6].cxt=2048/comp[6].a; } } // 7 MIX 16 0 7 24 255 { int err=(y*32767-squash(p[7]))*24>>4; int* wt=(int*)&comp[7].cm[comp[7].cxt]; wt[0]=clamp512k(wt[0]+((err*p[0]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err*p[1]+(1<<12))>>13)); wt[2]=clamp512k(wt[2]+((err*p[2]+(1<<12))>>13)); wt[3]=clamp512k(wt[3]+((err*p[3]+(1<<12))>>13)); wt[4]=clamp512k(wt[4]+((err*p[4]+(1<<12))>>13)); wt[5]=clamp512k(wt[5]+((err*p[5]+(1<<12))>>13)); wt[6]=clamp512k(wt[6]+((err*p[6]+(1<<12))>>13)); } break; } // max.cfg case 3: { // 22 components // 0 CONST 160 // 1 ICM 5 { comp[1].ht[comp[1].c+(hmap4&15)]= st.next(comp[1].ht[comp[1].c+(hmap4&15)], y); U32& pn=comp[1].cm(comp[1].cxt); pn+=int(y*32767-(pn>>8))>>2; } // 2 ISSE 13 1 { int err=y*32767-squash(p[2]); int *wt=(int*)&comp[2].cm[comp[2].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[1]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[2].ht[comp[2].c+(hmap4&15)]=st.next(comp[2].cxt, y); } // 3 ISSE 16 2 { int err=y*32767-squash(p[3]); int *wt=(int*)&comp[3].cm[comp[3].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[2]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[3].ht[comp[3].c+(hmap4&15)]=st.next(comp[3].cxt, y); } // 4 ISSE 18 3 { int err=y*32767-squash(p[4]); int *wt=(int*)&comp[4].cm[comp[4].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[3]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[4].ht[comp[4].c+(hmap4&15)]=st.next(comp[4].cxt, y); } // 5 ISSE 19 4 { int err=y*32767-squash(p[5]); int *wt=(int*)&comp[5].cm[comp[5].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[4]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[5].ht[comp[5].c+(hmap4&15)]=st.next(comp[5].cxt, y); } // 6 ISSE 19 5 { int err=y*32767-squash(p[6]); int *wt=(int*)&comp[6].cm[comp[6].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[5]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[6].ht[comp[6].c+(hmap4&15)]=st.next(comp[6].cxt, y); } // 7 ISSE 20 6 { int err=y*32767-squash(p[7]); int *wt=(int*)&comp[7].cm[comp[7].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[6]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[7].ht[comp[7].c+(hmap4&15)]=st.next(comp[7].cxt, y); } // 8 MATCH 22 24 { if (comp[8].c!=y) comp[8].a=0; comp[8].ht(comp[8].limit>>3)+=comp[8].ht(comp[8].limit>>3)+y; if ((++comp[8].limit&7)==0) { int pos=comp[8].limit>>3; if (comp[8].a==0) { comp[8].b=pos-comp[8].cm(z.H(8)); if (comp[8].b&(comp[8].ht.size()-1)) while (comp[8].a<255 && comp[8].ht(pos-comp[8].a-1) ==comp[8].ht(pos-comp[8].a-comp[8].b-1)) ++comp[8].a; } else comp[8].a+=comp[8].a<255; comp[8].cm(z.H(8))=pos; if (comp[8].a>0) comp[8].cxt=2048/comp[8].a; } } // 9 ICM 17 { comp[9].ht[comp[9].c+(hmap4&15)]= st.next(comp[9].ht[comp[9].c+(hmap4&15)], y); U32& pn=comp[9].cm(comp[9].cxt); pn+=int(y*32767-(pn>>8))>>2; } // 10 ISSE 19 9 { int err=y*32767-squash(p[10]); int *wt=(int*)&comp[10].cm[comp[10].cxt*2]; wt[0]=clamp512k(wt[0]+((err*p[9]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err+16)>>5)); comp[10].ht[comp[10].c+(hmap4&15)]=st.next(comp[10].cxt, y); } // 11 ICM 13 { comp[11].ht[comp[11].c+(hmap4&15)]= st.next(comp[11].ht[comp[11].c+(hmap4&15)], y); U32& pn=comp[11].cm(comp[11].cxt); pn+=int(y*32767-(pn>>8))>>2; } // 12 ICM 13 { comp[12].ht[comp[12].c+(hmap4&15)]= st.next(comp[12].ht[comp[12].c+(hmap4&15)], y); U32& pn=comp[12].cm(comp[12].cxt); pn+=int(y*32767-(pn>>8))>>2; } // 13 ICM 13 { comp[13].ht[comp[13].c+(hmap4&15)]= st.next(comp[13].ht[comp[13].c+(hmap4&15)], y); U32& pn=comp[13].cm(comp[13].cxt); pn+=int(y*32767-(pn>>8))>>2; } // 14 ICM 14 { comp[14].ht[comp[14].c+(hmap4&15)]= st.next(comp[14].ht[comp[14].c+(hmap4&15)], y); U32& pn=comp[14].cm(comp[14].cxt); pn+=int(y*32767-(pn>>8))>>2; } // 15 MIX 16 0 15 24 255 { int err=(y*32767-squash(p[15]))*24>>4; int* wt=(int*)&comp[15].cm[comp[15].cxt]; wt[0]=clamp512k(wt[0]+((err*p[0]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err*p[1]+(1<<12))>>13)); wt[2]=clamp512k(wt[2]+((err*p[2]+(1<<12))>>13)); wt[3]=clamp512k(wt[3]+((err*p[3]+(1<<12))>>13)); wt[4]=clamp512k(wt[4]+((err*p[4]+(1<<12))>>13)); wt[5]=clamp512k(wt[5]+((err*p[5]+(1<<12))>>13)); wt[6]=clamp512k(wt[6]+((err*p[6]+(1<<12))>>13)); wt[7]=clamp512k(wt[7]+((err*p[7]+(1<<12))>>13)); wt[8]=clamp512k(wt[8]+((err*p[8]+(1<<12))>>13)); wt[9]=clamp512k(wt[9]+((err*p[9]+(1<<12))>>13)); wt[10]=clamp512k(wt[10]+((err*p[10]+(1<<12))>>13)); wt[11]=clamp512k(wt[11]+((err*p[11]+(1<<12))>>13)); wt[12]=clamp512k(wt[12]+((err*p[12]+(1<<12))>>13)); wt[13]=clamp512k(wt[13]+((err*p[13]+(1<<12))>>13)); wt[14]=clamp512k(wt[14]+((err*p[14]+(1<<12))>>13)); } // 16 MIX 8 0 16 10 255 { int err=(y*32767-squash(p[16]))*10>>4; int* wt=(int*)&comp[16].cm[comp[16].cxt]; wt[0]=clamp512k(wt[0]+((err*p[0]+(1<<12))>>13)); wt[1]=clamp512k(wt[1]+((err*p[1]+(1<<12))>>13)); wt[2]=clamp512k(wt[2]+((err*p[2]+(1<<12))>>13)); wt[3]=clamp512k(wt[3]+((err*p[3]+(1<<12))>>13)); wt[4]=clamp512k(wt[4]+((err*p[4]+(1<<12))>>13)); wt[5]=clamp512k(wt[5]+((err*p[5]+(1<<12))>>13)); wt[6]=clamp512k(wt[6]+((err*p[6]+(1<<12))>>13)); wt[7]=clamp512k(wt[7]+((err*p[7]+(1<<12))>>13)); wt[8]=clamp512k(wt[8]+((err*p[8]+(1<<12))>>13)); wt[9]=clamp512k(wt[9]+((err*p[9]+(1<<12))>>13)); wt[10]=clamp512k(wt[10]+((err*p[10]+(1<<12))>>13)); wt[11]=clamp512k(wt[11]+((err*p[11]+(1<<12))>>13)); wt[12]=clamp512k(wt[12]+((err*p[12]+(1<<12))>>13)); wt[13]=clamp512k(wt[13]+((err*p[13]+(1<<12))>>13)); wt[14]=clamp512k(wt[14]+((err*p[14]+(1<<12))>>13)); wt[15]=clamp512k(wt[15]+((err*p[15]+(1<<12))>>13)); } // 17 MIX2 0 15 16 24 0 { int err=(y*32767-squash(p[17]))*24>>5; int w=comp[17].a16[comp[17].cxt]; w+=(err*(p[15]-p[16])+(1<<12))>>13; if (w<0) w=0; if (w>65535) w=65535; comp[17].a16[comp[17].cxt]=w; } // 18 SSE 8 17 32 255 train(comp[18], y); // 19 MIX2 8 17 18 16 255 { int err=(y*32767-squash(p[19]))*16>>5; int w=comp[19].a16[comp[19].cxt]; w+=(err*(p[17]-p[18])+(1<<12))>>13; if (w<0) w=0; if (w>65535) w=65535; comp[19].a16[comp[19].cxt]=w; } // 20 SSE 16 19 32 255 train(comp[20], y); // 21 MIX2 0 19 20 16 0 { int err=(y*32767-squash(p[21]))*16>>5; int w=comp[21].a16[comp[21].cxt]; w+=(err*(p[19]-p[20])+(1<<12))>>13; if (w<0) w=0; if (w>65535) w=65535; comp[21].a16[comp[21].cxt]=w; } break; } // Not optimized default: return update0(y); } c8+=c8+y; if (c8>=256) { z.run(c8-256); hmap4=1; c8=1; } else if (c8>=16 && c8<32) hmap4=(hmap4&0xf)<<5|y<<4|1; else hmap4=(hmap4&0x1f0)|(((hmap4&0xf)*2+y)&0xf); } void ZPAQL::run(U32 input) { switch(select) { // fast.cfg case 1: { a = input; m(b) = a; a = 0; d = 0; a = (a+m(b)+512)*773; --b; a = (a+m(b)+512)*773; h(d) = a; ++d; --b; a = (a+m(b)+512)*773; --b; a = (a+m(b)+512)*773; h(d) = a; return; break; } // mid.cfg case 2: { a = input; ++c; m(c) = a; b = c; a = 0; d = 1; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; --b; a = (a+m(b)+512)*773; h(d) = a; ++d; a = m(c); a <<= (8&31); h(d) = a; return; break; } // max.cfg case 3: { a = input; ++c; m(c) = a; b = c; a = 0; d = 2; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; --b; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = (a+m(b)+512)*773; h(d) = a; --b; ++d; a = m(c); a &= ~ 32; f = (a > U32(64)); if (!f) goto L300057; f = (a < U32(91)); if (!f) goto L300057; ++d; h(d) = (h(d)+a+512)*773; --d; swap(h(d)); a += h(d); a *= 20; h(d) = a; goto L300066; L300057: a = h(d); f = (a == U32(0)); if (f) goto L300065; ++d; h(d) = a; --d; L300065: h(d) = 0; L300066: ++d; ++d; b = c; --b; a = 0; a = (a+m(b)+512)*773; h(d) = a; ++d; --b; a = 0; a = (a+m(b)+512)*773; h(d) = a; ++d; --b; a = 0; a = (a+m(b)+512)*773; h(d) = a; ++d; a = b; a -= 212; b = a; a = 0; a = (a+m(b)+512)*773; h(d) = a; swap(b); a -= 216; swap(b); a = m(b); a &= 60; h(d) = (h(d)+a+512)*773; ++d; a = m(c); a <<= (9&31); h(d) = a; ++d; ++d; ++d; ++d; ++d; h(d) = a; return; break; } // Not optimized default: run0(input); } } ////////////////////////////// Decoder //////////////////////////// // Decoder decompresses using an arithmetic code class Decoder { FILE* in; // destination U32 low, high; // range U32 curr; // last 4 bytes of archive Predictor pr; // to get p int decode(int p); // return decoded bit (0..1) with probability p (0..8191) public: Decoder(FILE* f, ZPAQL& z); int decompress(); // return a byte or EOF int skip(); // skip to the end of the segment, return next byte }; Decoder::Decoder(FILE* f, ZPAQL& z): in(f), low(1), high(0xFFFFFFFF), curr(0), pr(z) {} inline int Decoder::decode(int p) { assert(p>=0 && p<65536); assert(high>low && low>0); if (currhigh) error("archive corrupted"); assert(curr>=low && curr<=high); U32 mid=low+((high-low)>>16)*p+((((high-low)&0xffff)*p)>>16); // split range assert(high>mid && mid>=low); int y=curr<=mid; if (y) high=mid; else low=mid+1; // pick half while ((high^low)<0x1000000) { // shift out identical leading bytes high=high<<8|255; low=low<<8; low+=(low==0); int c=getc(in); if (c==EOF) error("unexpected end of file"); curr=curr<<8|c; } return y; } int Decoder::decompress() { if (curr==0) { // finish initialization for (int i=0; i<4; ++i) curr=curr<<8|getc(in); } if (decode(0)) { if (curr!=0) error("decoding end of stream"); return EOF; } else { int c=1; while (c<256) { // get 8 bits int p=pr.predict()*2+1; c+=c+decode(p); pr.update(c&1); } return c-256; } } // Find end of compressed data and return next byte int Decoder::skip() { int c=0; while (curr==0) // at start? curr=getc(in); while (curr && (c=getc(in))!=EOF) // find 4 zeros curr=curr<<8|c; while ((c=getc(in))==0) ; // might be more than 4 return c; } /////////////////////////// PostProcessor //////////////////// class PostProcessor { int state; // input parse state int hsize; // header size int ph, pm; // sizes of H and M in z public: ZPAQL z; // holds PCOMP PostProcessor(ZPAQL& hz); void set(FILE* out, SHA1* p) {z.output=out; z.sha1=p;} // Set output int write(int c); // Input a byte, return state }; // Copy ph, pm from block header PostProcessor::PostProcessor(ZPAQL& hz) { state=hsize=0; ph=hz.header[4]; pm=hz.header[5]; } // (PASS=0 | PROG=1 psize[0..1] pcomp[0..psize-1]) data... EOB=-1 // Return state: 1=PASS, 2..4=loading PROG, 5=PROG loaded int PostProcessor::write(int c) { assert(c>=-1 && c<=255); switch (state) { case 0: // initial state if (c<0) error("Unexpected EOS"); state=c+1; // 1=PASS, 2=PROG if (state>2) error("unknown post processing type"); break; case 1: // PASS if (z.output && c>=0) putc(c, z.output); // data if (z.sha1 && c>=0) z.sha1->put(c); break; case 2: // PROG if (c<0) error("Unexpected EOS"); hsize=c; // low byte of size state=3; break; case 3: // PROG psize[0] if (c<0) error("Unexpected EOS"); hsize+=c*256; // high byte of psize z.header.resize(hsize+300); z.cend=8; z.hbegin=z.hend=z.cend+128; z.header[4]=ph; z.header[5]=pm; state=4; break; case 4: // PROG psize[0..1] pcomp[0...] if (c<0) error("Unexpected EOS"); assert(z.hend>8; z.initp(); z.selectModel(0); state=5; } break; case 5: // PROG ... data z.run(c); break; } return state; } /////////////////////////// Decompress /////////////////////// // Open archive. Append .zpaq to file name if missing. // filename and mode are as in fopen(). Error if cannot open. FILE *open_archive(const char *filename, const char *mode) { assert(filename); assert(mode); int len=strlen(filename); Array newname(len+6); append(newname, filename); if (len<5 || strcmp(filename+len-5, ".zpaq")) append(newname, ".zpaq"); FILE *f=fopen(&newname[0], mode); if (!f) perror(&newname[0]), error("cannot open archive"); switch(mode[0]) { case 'r': printf("Reading from archive %s\n", &newname[0]); break; case 'w': printf("Created archive %s\n", &newname[0]); break; case 'a': printf("Appending to archive %s\n", &newname[0]); break; } return f; } // Reject archive filenames with absolute paths, drive letters // or control characters or that are too long. static bool validate_filename(const char* filename) { int len=strlen(filename); if (len<1) return true; // No name is OK if (len>511) return false; // name too long if (strstr(filename, "../")) return false; // no backward paths if (strstr(filename, "..\\")) return false; if (filename[0]=='/' || filename[0]=='\\') return false; // no absolute path for (int i=0; iLEVEL || c<1 || getc(in)!=1) error("not ZPAQ"); // Skip block header int hsize=getc(in); hsize+=getc(in)*256; if (hsize<6 || hsize>65535) error("hsize missing"); while (hsize-->0) getc(in); // Skip segments while ((c=getc(in))==1) { ++segments; while (getc(in)>0) ; // skip filename while (getc(in)>0) ; // skip comment if (getc(in)!=0) error("reserved 0 missing"); // Skip to end of data U32 c4=0xFFFFFFFF; // last 4 bytes will be all 0 while ((c=getc(in))!=EOF && (c4=c4<<8|c)!=0) ; if (c==EOF) error("unexpected end of file"); while ((c=getc(in))==0) ; if (c==253) { // Skip SHA1 for (int i=0; i<20; ++i) getc(in); } else if (c!=254) error("missing end of segment marker"); } if (c!=255) error("missing end of block marker"); return segments; } // Remove path from filename static char* strip(char* filename) { assert(filename); int len=strlen(filename); char *result=filename; for (int i=0; i0 if Linux, 0 if unknown if (os==0) { for (int i=0; path && path[i]; ++i) { if (path[i]=='/') ++os; if (path[i]=='\\') --os; } } // Change slashes in filename per OS if known. for (int i=0; filename[i]; ++i) { if (os>0 && filename[i]=='\\') filename[i]='/'; if (os<0 && filename[i]=='/') filename[i]='\\'; } // Try opening file FILE *f=fopen(filename, "wb"); if (f) return f; // If this doesn't work, try creating a directory for it using "mkdir" if (os && errno==ENOENT) { Array cmd(slash+16); strcpy(&cmd[0], os<=0 ? "mkdir " : "mkdir -p "); strncat(&cmd[0], filename, slash); printf("%s\n", &cmd[0]); system(&cmd[0]); // Last try return fopen(filename, "wb"); } return 0; } // Decompress: eN|xN archive [files...] static void decompress(int argc, char** argv) { assert(argc>=3); // Open archive FILE* in=open_archive(argv[2], "rb"); // If user specifies N then skip N-1 blocks int block=atoi(argv[1]+1); if (block>0) { for (; block>1; --block) skip_block(in); } // Read the archive int filecount=0; // number of files extracted FILE *out=0; // output file int c; while (find_start(in)) { if (getc(in)!=LEVEL || getc(in)!=1) error("Not ZPAQ"); // Read block header ZPAQL z; z.read(Reader(in)); // PostProcessor and Decoder is created and and destroyed for each block PostProcessor pp(z); Decoder dec(in, z); // Read segments bool first=true; // first segment of block? while ((c=getc(in))==1) { // Read the filename char filename[512]={0}; int i; for (i=0; (c=getc(in))>0; ++i) if (i<511) filename[i]=c; if (i>0 && i<512) filename[i]=0; printf("%s ", filename); // Get comment char comment[20]={0}; i=0; while ((c=getc(in))!=EOF && c!=0) { if (i<19) comment[i]=c; ++i; } printf("%s -> ", comment); if (getc(in)) error("reserved"); // reserved 0 // open output file // if filename is empty, use the previously opened file if (filename[0] || !out) { // close last file if (out) { fclose(out); out=0; ++filecount; } // if the user gave an output file starting at argv[3], use it instead. if (argc>3) { if (filecount+3>=argc) { printf("and remaining files not extracted\n"); goto end; } char* name=argv[filecount+3]; out=create(name); if (!out) { perror(name); goto end; } else printf("%s ", name); } // Otherwise, use the names in the archive, but don't clobber // or use suspicious filenames else { char* newname=filename; if (argv[1][0]=='e') newname=strip(filename); if (newname!=filename) printf("%s -> ", newname); if (!validate_filename(newname)) { printf("Error: bad filename\n"); goto end; } out=fopen(newname, "rb"); if (out) { fclose(out); out=0; printf("Error: won't overwrite\n"); goto end; } else { out=create(newname); if (!out) { perror(newname); goto end; } } } } // Decompress SHA1 sha1; pp.set(out, &sha1); // Extract the current segment { time_t now=time(0); int len=0; while ((c=dec.decompress())!=EOF) { if (pp.write(c)==5 && first) { first=false; } if (!(len++&0xfff) && time(0)!=now) { for (int i=printf("%1.0f ", sha1.size()); i>0; --i) putchar('\b'); fflush(stdout); now=time(0); } } pp.write(-1); } // Check for end of segment and block markers int eos=c; eos=getc(in); // 253=SHA1 follows, 254=EOS if (eos==253) { U8 hash[20]; bool match=true; for (int i=0; i<20; ++i) { hash[i]=getc(in); if (hash[i]!=sha1.result(i)) match=false; } if (1) { if (match) { printf("Checksum OK "); } else { fprintf(stderr, "CHECKSUM FAILED: FILE IS NOT IDENTICAL\n Archive SHA1: "); for (int i=0; i<20; ++i) fprintf(stderr, "%02x", hash[i]); fprintf(stderr, "\n File SHA1: "); for (int i=0; i<20; ++i) fprintf(stderr, "%02x", sha1.result(i)); fprintf(stderr, "\n"); } } } else if (eos!=254) error("missing end of segment marker"); else printf("OK, no checksum "); printf("\n"); } if (c!=255) error("missing end of block marker"); if (block) break; } // Close files end: if (out) fclose(out), ++filecount; fclose(in); printf("%d file(s) extracted\n", filecount); } //////////////////////////// Encoder /////////////////////////////// // Encoder compresses using an arithmetic code class Encoder { FILE* out; // destination U32 low, high; // range Predictor pr; // to get p void encode(int y, int p); // encode bit y (0..1) with probability p (0..8191) U32 in_low, in_high; // number of input, output bytes (64 bits) U32 out_low, out_high; public: Encoder(FILE* f, ZPAQL& z); void compress(int c); // c is 0..255 or EOF // void stat() {pr.stat();} // print predictor statistics void setOutput(FILE* f) {out=f;} double in_size() const {return in_low+4294967296.0*in_high;} double out_size() const {return out_low+4294967296.0*out_high;} void reset() {in_low=in_high=out_low=out_high=0;} // clear sizes }; // Compress to file f using model z Encoder::Encoder(FILE* f, ZPAQL& z): out(f), low(1), high(0xFFFFFFFF), pr(z) { reset(); } // compress bit y having probability p/64K inline void Encoder::encode(int y, int p) { assert(out); assert(p>=0 && p<65536); assert(y==0 || y==1); assert(high>low && low>0); U32 mid=low+((high-low)>>16)*p+((((high-low)&0xffff)*p)>>16); // split range assert(high>mid && mid>=low); if (y) high=mid; else low=mid+1; // pick half while ((high^low)<0x1000000) { // write identical leading bytes putc(high>>24, out); // same as low>>24 high=high<<8|255; low=low<<8; low+=(low==0); // so we don't code 4 0 bytes in a row out_high+=(++out_low==0); } } // compress byte c (0..255 or -1=EOS) void Encoder::compress(int c) { assert(out); if (c==-1) encode(1, 0); else { assert(c>=0 && c<=255); in_high+=(++in_low==0); encode(0, 0); for (int i=7; i>=0; --i) { int p=pr.predict()*2+1; assert(p>0 && p<65536); int y=c>>i&1; encode(y, p); pr.update(y); } } } //////////////////////////// Compress //////////////////////////// // Test for regular file (Linux) static bool is_file(const char* filename) { #ifdef unix struct stat st; return stat(filename, &st)==0 && (st.st_mode & S_IFREG); #endif return true; } // Compress files: c|a archive files... static void compress(int argc, char** argv) { assert(argc>=3); ZPAQL z, pz; // compression and postprocessing models // Select compression option int sel=atoi(argv[1]+1); if (sel<1) sel=2; z.selectModel(sel); // Compress files in argv[3...argc-1] FILE *out=0; // archive opened when ready to compress first file Encoder enc(out, z); // compressor double outsum=0; // total output size for (int i=3; i=0 && psize<0x10000); assert(pz.header.size()>=pz.hend); if (psize==0) enc.compress(0); // PASS else { enc.compress(1); // POST enc.compress(psize&255); // size low byte enc.compress(psize>>8&255); // size high byte for (int j=0; j %1.0f ", presize); int len=0; time_t now=time(0); while ((c=getc(in))!=EOF) { enc.compress(c); if (!(len++&0xfff) && now!=time(0)) { for (int j=printf("%1.0f -> %1.0f ", enc.in_size(), outsize+enc.out_size()); j>0; --j) putchar('\b'); fflush(stdout); now=time(0); } } enc.compress(-1); // Write segment trailer outsize+=20+fprintf(out, "%c%c%c%c%c", 0, 0, 0, 0, 253); for (int j=0; j<20; ++j) putc(check1.result(j), out); fclose(in); in=0; printf("-> %1.0f \n", outsize+enc.out_size()); outsum+=outsize+enc.out_size(); } // Code end of block and close archive if (out) { putc(255, out); // block trailer printf("-> %1.0f\n", outsum); fclose(out); } else printf("Archive %s not updated\n", argv[2]); } ////////////////////////// list ////////////////////////// // List archive contents: l archive static void list(int argc, char** argv) { assert(argc>2 && argv[2]); // Open archive FILE* in=open_archive(argv[2], "rb"); // Read the file int c, blocks=0; while (find_start(in)) { // Read block header if (getc(in)!=LEVEL || getc(in)!=1) error("not ZPAQ"); ZPAQL z; double size=6+z.read(in); // compressed size printf("Block %d: compressed with option %d, requires %1.3f MB memory\n", ++blocks, z.select, z.memory()/1000000); // Read segments while ((c=getc(in))==1) { // Print filename and comments printf(" "); while ((c=getc(in))!=EOF && c) putchar(c), size+=1; printf(" "); while ((c=getc(in))!=EOF && c) putchar(c), size+=1; if (getc(in)!=0) error("reserved data"); size+=6; // Skip to end of data U32 c4=0xFFFFFFFF; // last 4 bytes will be all 0 while ((c=getc(in))!=EOF && (c4=c4<<8|c)!=0) size+=1; if (c==EOF) error("unexpected end of file"); while ((c=getc(in))==0) size+=1; if (c==253) { // print SHA1 printf(" SHA1="); size+=20; for (int i=0; i<20; ++i) { int c=getc(in); if (i<4) printf("%02x", c); } printf("..."); } else if (c!=254) error("missing end of segment marker"); printf(" -> %1.0f\n", size); size=0; } if (c!=255) error("missing end of block marker"); } } ///////////////////////////// Main /////////////////////////// // Print help message and exit static void usage() { printf("ZP v1.00 archiver, (C) 2010, Ocarina Networks Inc.\n" "Written by Matt Mahoney, " __DATE__ ".\n" "Licensed under GPL v3, http://www.gnu.org/copyleft/gpl.html\n" "\n" "Usage: zp command archive.zpaq [files...]\n" "Commands:\n" " l List archive contents\n" " x Extract with full path names (files... overrides stored names)\n" " e Extract to current directory\n" " xN, eN Extract only block N (1, 2, 3...)\n" " c Create new archive\n" " a Append to archive\n" " cN, aN Compress with option N\n" "Compression options:\n" " 1,2,3 Fast, medium, small (default is 2)\n" ); exit(0); } // Command syntax as in usage() int main(int argc, char** argv) { // Check usage if (argc<2) usage(); // Do the command char cmd=argv[1][0]; if (argc>=4 && (cmd=='a' || cmd=='c')) compress(argc, argv); else if (argc>=3 && (cmd=='x' || cmd=='e')) decompress(argc, argv); else if (argc>=3 && cmd=='l') list(argc, argv); else usage(); // Print time used printf("Elapsed time %1.2f seconds.\n", double(clock())/CLOCKS_PER_SEC); return 0; }