From 9b158e1280f94bc86e6b6a8914c52738e9cf483b Mon Sep 17 00:00:00 2001 From: zvezdochiot Date: Thu, 4 Aug 2022 09:38:22 +0300 Subject: [PATCH] 1.0: init --- LICENSE | 674 +++++++++++++++++++++++++++ Makefile | 24 + README.md | 240 ++++++++++ src/paq9a.cpp | 1222 +++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 2160 insertions(+) create mode 100644 LICENSE create mode 100644 Makefile create mode 100644 README.md create mode 100644 src/paq9a.cpp diff --git a/LICENSE b/LICENSE new file mode 100644 index 0000000..94a9ed0 --- /dev/null +++ b/LICENSE @@ -0,0 +1,674 @@ + 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 + + How to Apply These Terms to Your New Programs + + If you develop a new program, and you want it to be of the greatest +possible use to the public, the best way to achieve this is to make it +free software which everyone can redistribute and change under these terms. + + To do so, attach the following notices to the program. It is safest +to attach them to the start of each source file to most effectively +state the exclusion of warranty; and each file should have at least +the "copyright" line and a pointer to where the full notice is found. + + + Copyright (C) + + 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. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . + +Also add information on how to contact you by electronic and paper mail. + + If the program does terminal interaction, make it output a short +notice like this when it starts in an interactive mode: + + Copyright (C) + This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. + This is free software, and you are welcome to redistribute it + under certain conditions; type `show c' for details. + +The hypothetical commands `show w' and `show c' should show the appropriate +parts of the General Public License. Of course, your program's commands +might be different; for a GUI interface, you would use an "about box". + + You should also get your employer (if you work as a programmer) or school, +if any, to sign a "copyright disclaimer" for the program, if necessary. +For more information on this, and how to apply and follow the GNU GPL, see +. + + The GNU General Public License does not permit incorporating your program +into proprietary programs. If your program is a subroutine library, you +may consider it more useful to permit linking proprietary applications with +the library. If this is what you want to do, use the GNU Lesser General +Public License instead of this License. But first, please read +. diff --git a/Makefile b/Makefile new file mode 100644 index 0000000..a40c724 --- /dev/null +++ b/Makefile @@ -0,0 +1,24 @@ +TARGET = paq9a +CXX ?= g++ +CXXFLAGS ?= -Wall -O3 +LDFLAGS ?= -s +AR = ar +RM ?= rm -f +SRCS = src/$(TARGET).cpp +OBJS = $(SRCS:%.cpp=%.o) +LDLIBS = + +ifeq ($(STATIC), Y) + LDFLAGS += -static +endif + +all: $(TARGET) + +$(TARGET): $(OBJS) + $(CXX) $(CXXFLAGS) $^ $(LDFLAGS) $(LDLIBS) -o $@ + +%.o: %.cpp + $(CXX) $(CXXFLAGS) -c $< -o $@ + +clean: + $(RM) $(TARGET) $(OBJS) diff --git a/README.md b/README.md new file mode 100644 index 0000000..8b849c4 --- /dev/null +++ b/README.md @@ -0,0 +1,240 @@ +`ORG.FSMS:` +![GitHub release (latest by date)](https://img.shields.io/github/v/release/FS-make-simple/paq9a) +![GitHub Release Date](https://img.shields.io/github/release-date/FS-make-simple/paq9a) +![GitHub repo size](https://img.shields.io/github/repo-size/FS-make-simple/paq9a) +![GitHub all releases](https://img.shields.io/github/downloads/FS-make-simple/paq9a/total) +![GitHub](https://img.shields.io/github/license/FS-make-simple/paq9a) + +# paq9a archiver +Dec. 31, 2007 (C) 2007, Matt Mahoney + +## 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 . + +## Intro + +`paq9a` is an experimental file compressor and archiver. + +## Usage +```sh + paq9a {a|x|l} archive [[-opt] files...]... +``` +Commands: +``` + a = create archive and compress named files. + x = extract from archive. + l = list contents. +``` +Archives are "solid". You can only create new archives. You cannot +modify existing archives. File names are stored and extracted exactly as +named when the archive is created, but you have the option to rename them +during extraction. Files are never clobbered. + +The "a" command creates a new archive and adds the named files. +Wildcards are permitted if compiled with g++. Options +and filenames may be in any order. Options apply only to filenames +after the option, and override previous options. +Options are: +``` + -s = store without compression. + -c = compress (default). + -1 through -9 selects memory level from 18 MB to 1.5 GB Default is -7 + using 405 MB. The memory option must be set before the first file. + Decompression requires the same amount of memory. +``` +For example: +``` + paq9a a foo.paq9a a.txt -3 -s b.txt -c c.txt tmp/d.txt /tmp/e.txt +``` +creates the archive foo.paq9a with 5 files. The file b.txt is +stored without compression. The other 4 files are compressed +at memory level 3. Extraction requires the same memory as compression. + +If any named file does not exist, then it is omitted from the archive +with a warning and the remaining files are added. An existing +archive cannot be overwritten. There must be at least one filename on +the command line. + +The "x" command extracts the archive contents, creating files exactly +as named when the archive was created. Files cannot be overwritten. +If a file already exists or cannot be created, then it is skipped. +For example, "tmp/d.txt" would be skipped if either the current +directory does not have a subdirectory tmp, or tmp is write +protected, or tmp/d.txt already exists. + +If "x" is followed by one or more file names, then the output files +are renamed in the order they were added to the archive and any remaining +contents are extracted without renaming. +For example: +```sh + paq9a x foo.paq9a x.txt y.txt +``` +would extract a.txt to x.txt and b.txt to y.txt, then extract c.txt, +tmp/d.txt and /tmp/e.txt. If the command line has more filenames than +the archive then the extra arguments are ignored. Options are not +allowed. + +The "l" (letter l) command lists the contents. Any extra arguments +are ignored. + +Any other command, or no command, displays a help message. + +## ARCHIVE FORMAT +``` + "lPq" 1 mem [filename {'\0' mode usize csize contents}...]... +``` +The first 4 bytes are "lPq\x01" (1 is the version number). + +`mem` is a digit '1' through '9', where '9' uses the most memory (1.5 GB). + +A file is stored as one or more blocks. The filename is stored +only in the first block as a NUL terminated string. Subsequent +blocks start with a 0. + +The mode is 's' if the block is stored and 'c' if compressed. + +`usize` = uncompressed size as a 4 byte big-endian number (MSB first). + +`csize` = compressed size as a 4 byte big-endian number. + +The contents is copied from the file itself if mode is 's' or the +compressed contents otherwise. Its length is exactly csize bytes. + +## COMPRESSED FORMAT + +Files are preprocessed with LZP and then compressed with a context +mixing compressor and arithmetic coded one bit at a time. Model +contents are maintained across files. + +The LZP stage predicts the next byte by matching the current context +(order 12 or higher) to a rotating buffer. If a match is found +then the next byte after the match is predicted. If the next byte +matches the prediction, then a 1 bit is coded and the context is extended. +Otherwise a 0 is coded followed by 8 bits of the actual byte in MSB to +LSB order. + +A 1 bit is modeled using the match length as context, then refined +in 3 stages using sucessively longer contexts. The predictions are +adjusted by 2 input neurons selected by a context hash with the second +input fixed. + +If the LZP prediction is missed, then the literal is coded using a chain +of predicions which are mixed using neurons, where one input is the +previous prediction and the second input is the prediction given the +current context. The current context is mapped to an 8 bit state +representing the bit history, the sequence of bits previously observed +in that context. The bit history is used both to select the neuron +and is mapped to a prediction that provides the second input. In addition, +if the known bits of the current byte match the LZP incorrectly predicted +byte, then this fact is used to select one of 2 sets of neurons (512 total). + +The contexts, in order, are sparse order-1 with gaps of 3, 2, and 1 +byte, then orders 1 through 6, then word orders 0 and 1, where a word +is a sequenece of case insensitive letters (useful for compressing text). +Contexts longer than 1 are hashed. Order-n contexts consist of a hash +of the last n bytes plus the 0 to 7 known bits of the current byte. +The order 6 context and the word order 0 and 1 contexts also include +the LZP predicted byte. + +All mixing is in the logistic or "stretched" domain: stretch(p) = ln(p/(1-p)), +then "squashed" by the inverse function: squash(p) = 1/(1 + exp(-p)) before +arithmetic coding. A 2 input neuron has 2 weights (w0 and w1) +selected by context. Given inputs x0 and x1 (2 predictions, or one +prediction and a constant), the output prediction is computed: +p = w0*x0 + w1*x1. If the actual bit is y, then the weights are updated +to minimize its coding cost: +``` + error = y - squash(p) + w0 += x0 * error * L + w1 += x1 * error * L +``` +where L is the learning rate, normally 1/256, but increased by a factor +of 4 an 2 for the first 2 training cycles (using the 2 low bits +of w0 as a counter). In the implementation, p is represented by a fixed +point number with a 12 bit fractional part in the linear domain (0..4095) +and 8 bits in the logistic domain (-2047..2047 representing -8..8). +Weights are scaled by 24 bits. Both weights are initialized to 1/2, +expecting 2 probabilities, weighted equally). However, when one input +(x0) is fixed, its weight (w0) is initialized to 0. + +A bit history represents the sequence of 0 and 1 bits observed in a given +context. An 8 bit state represents all possible sequences up to 4 bits +long. Longer sequences are represented by a count of 0 and 1 bits, plus +an indicator of the most recent bit. If counts grow too large, then the +next state represents a pair of smaller counts with about the same ratio. +The state table is the same as used in PAQ8 (all versions) and LPAQ1. + +A state is mapped to a prediction by using a table. A table entry +contains 2 values, p, initialized to 1/2, and n, initialized to 0. +The output prediciton is p (in the linear domain, not stretched). +If the actual bit is y, then the entry is updated: +``` + error = y - p + p += error/(n + 1.5) + if n < limit then n += 1 +``` +In practice, p is scaled by 22 bits, and n is 10 bits, packed into +one 32 bit integer. The limit is 255. + +Every 4 bits, contexts are mapped to arrays of 15 states using a +hash table. The first element is the bit history for the current +context ending on a half byte boundary, followed by all possible +contexts formed by appending up to 3 more bits. + +A hash table accepts a 32 bit context, which must be a hash if +longer than 4 bytes. The input is further hashed and divided into +an index (depending on the table size, a power of 2), and an 8 bit +checksum which is stored in the table and used to detect collisions +(not perfectly). A lookup tests 3 adjacent locations within a single +64 byte cache line, and if a matching checksum is not found, then the +entry with the smallest value in the first data element is replaced +(LFU replacement policy). This element represents a bit history +for a context ending on a half byte boundary. The states are ordered +so that larger values represent larger total bit counts, which +estimates the likelihood of future use. The initial state is 0. + +Memory is allocated from MEM = pow(2, opt+22) bytes, where opt is 1 through +9 (user selected). Of this, MEM/2 is for the hash table for storing literal +context states, MEM/8 for the rotating LZP buffer, and MEM/8 for a +hash table of pointers into the buffer, plus 12 MB for miscellaneous data. +Total memory usage is 0.75*MEM + 12 MB. + +## ARITHMETIC CODING + +The arithmetic coder codes a bit with probability p using log2(1/p) bits. +Given input string y, the output is a binary fraction x such that +P(< y) <= x < P(<= y) where P(< y) means the total probability of all inputs +lexicographically less than y and P(<= y) = P(< y) + P(y). Note that one +can always find x with length at most log2(P(y)) + 1 bits. + +x can be computed efficiently by maintaining a range, low <= x < high +(initially 0..1) and expressing P(y) as a product of predictions: +P(y) = P(y1) P(y2|y1) P(y3|y1y2) P(y4|y1y2y3) ... P(yn|y1y2...yn-1) +where the term P(yi|y0y1...yi-1) means the probability that yi is 1 +given the context y1...yi-1, the previous i-1 bits of y. For each +prediction p, the range is split in proportion to the probabilities +of 0 and 1, then updated by taking the half corresponding to the actual +bit y as the new range, i.e. +``` + mid = low + (high - low) * p(y = 1) + if y = 0 then (low, high) := (mid, high) + if y = 1 then (low, high) := (low, mid) +``` +As low and high approach each other, the high order bits of x become +known (because they are the same throughout the range) and can be +output immediately. + +For decoding, the range is split as before and the range is updated +to the half containing x. The corresponding bit y is used to update +the model. Thus, the model has the same knowledge for coding and +decoding. diff --git a/src/paq9a.cpp b/src/paq9a.cpp new file mode 100644 index 0000000..58552af --- /dev/null +++ b/src/paq9a.cpp @@ -0,0 +1,1222 @@ +/* paq9a archiver, Dec. 31, 2007 (C) 2007, Matt Mahoney + + 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 . + +paq9a is an experimental file compressor and archiver. Usage: + + paq9a {a|x|l} archive [[-opt] files...]... + +Commands: + + a = create archive and compress named files. + x = extract from archive. + l = list contents. + +Archives are "solid". You can only create new archives. You cannot +modify existing archives. File names are stored and extracted exactly as +named when the archive is created, but you have the option to rename them +during extraction. Files are never clobbered. + +The "a" command creates a new archive and adds the named files. +Wildcards are permitted if compiled with g++. Options +and filenames may be in any order. Options apply only to filenames +after the option, and override previous options. Options are: + + -s = store without compression. + -c = compress (default). + -1 through -9 selects memory level from 18 MB to 1.5 GB Default is -7 + using 405 MB. The memory option must be set before the first file. + Decompression requires the same amount of memory. + +For example: + + paq9a a foo.paq9a a.txt -3 -s b.txt -c c.txt tmp/d.txt /tmp/e.txt + +creates the archive foo.paq9a with 5 files. The file b.txt is +stored without compression. The other 4 files are compressed +at memory level 3. Extraction requires the same memory as compression. + +If any named file does not exist, then it is omitted from the archive +with a warning and the remaining files are added. An existing +archive cannot be overwritten. There must be at least one filename on +the command line. + +The "x" command extracts the archive contents, creating files exactly +as named when the archive was created. Files cannot be overwritten. +If a file already exists or cannot be created, then it is skipped. +For example, "tmp/d.txt" would be skipped if either the current +directory does not have a subdirectory tmp, or tmp is write +protected, or tmp/d.txt already exists. + +If "x" is followed by one or more file names, then the output files +are renamed in the order they were added to the archive and any remaining +contents are extracted without renaming. For example: + + paq9a x foo.paq9a x.txt y.txt + +would extract a.txt to x.txt and b.txt to y.txt, then extract c.txt, +tmp/d.txt and /tmp/e.txt. If the command line has more filenames than +the archive then the extra arguments are ignored. Options are not +allowed. + +The "l" (letter l) command lists the contents. Any extra arguments +are ignored. + +Any other command, or no command, displays a help message. + + +ARCHIVE FORMAT + + "lPq" 1 mem [filename {'\0' mode usize csize contents}...]... + +The first 4 bytes are "lPq\x01" (1 is the version number). + +mem is a digit '1' through '9', where '9' uses the most memory (1.5 GB). + +A file is stored as one or more blocks. The filename is stored +only in the first block as a NUL terminated string. Subsequent +blocks start with a 0. + +The mode is 's' if the block is stored and 'c' if compressed. + +usize = uncompressed size as a 4 byte big-endian number (MSB first). + +csize = compressed size as a 4 byte big-endian number. + +The contents is copied from the file itself if mode is 's' or the +compressed contents otherwise. Its length is exactly csize bytes. + + +COMPRESSED FORMAT + +Files are preprocessed with LZP and then compressed with a context +mixing compressor and arithmetic coded one bit at a time. Model +contents are maintained across files. + +The LZP stage predicts the next byte by matching the current context +(order 12 or higher) to a rotating buffer. If a match is found +then the next byte after the match is predicted. If the next byte +matches the prediction, then a 1 bit is coded and the context is extended. +Otherwise a 0 is coded followed by 8 bits of the actual byte in MSB to +LSB order. + +A 1 bit is modeled using the match length as context, then refined +in 3 stages using sucessively longer contexts. The predictions are +adjusted by 2 input neurons selected by a context hash with the second +input fixed. + +If the LZP prediction is missed, then the literal is coded using a chain +of predicions which are mixed using neurons, where one input is the +previous prediction and the second input is the prediction given the +current context. The current context is mapped to an 8 bit state +representing the bit history, the sequence of bits previously observed +in that context. The bit history is used both to select the neuron +and is mapped to a prediction that provides the second input. In addition, +if the known bits of the current byte match the LZP incorrectly predicted +byte, then this fact is used to select one of 2 sets of neurons (512 total). + +The contexts, in order, are sparse order-1 with gaps of 3, 2, and 1 +byte, then orders 1 through 6, then word orders 0 and 1, where a word +is a sequenece of case insensitive letters (useful for compressing text). +Contexts longer than 1 are hashed. Order-n contexts consist of a hash +of the last n bytes plus the 0 to 7 known bits of the current byte. +The order 6 context and the word order 0 and 1 contexts also include +the LZP predicted byte. + +All mixing is in the logistic or "stretched" domain: stretch(p) = ln(p/(1-p)), +then "squashed" by the inverse function: squash(p) = 1/(1 + exp(-p)) before +arithmetic coding. A 2 input neuron has 2 weights (w0 and w1) +selected by context. Given inputs x0 and x1 (2 predictions, or one +prediction and a constant), the output prediction is computed: +p = w0*x0 + w1*x1. If the actual bit is y, then the weights are updated +to minimize its coding cost: + + error = y - squash(p) + w0 += x0 * error * L + w1 += x1 * error * L + +where L is the learning rate, normally 1/256, but increased by a factor +of 4 an 2 for the first 2 training cycles (using the 2 low bits +of w0 as a counter). In the implementation, p is represented by a fixed +point number with a 12 bit fractional part in the linear domain (0..4095) +and 8 bits in the logistic domain (-2047..2047 representing -8..8). +Weights are scaled by 24 bits. Both weights are initialized to 1/2, +expecting 2 probabilities, weighted equally). However, when one input +(x0) is fixed, its weight (w0) is initialized to 0. + +A bit history represents the sequence of 0 and 1 bits observed in a given +context. An 8 bit state represents all possible sequences up to 4 bits +long. Longer sequences are represented by a count of 0 and 1 bits, plus +an indicator of the most recent bit. If counts grow too large, then the +next state represents a pair of smaller counts with about the same ratio. +The state table is the same as used in PAQ8 (all versions) and LPAQ1. + +A state is mapped to a prediction by using a table. A table entry +contains 2 values, p, initialized to 1/2, and n, initialized to 0. +The output prediciton is p (in the linear domain, not stretched). +If the actual bit is y, then the entry is updated: + + error = y - p + p += error/(n + 1.5) + if n < limit then n += 1 + +In practice, p is scaled by 22 bits, and n is 10 bits, packed into +one 32 bit integer. The limit is 255. + +Every 4 bits, contexts are mapped to arrays of 15 states using a +hash table. The first element is the bit history for the current +context ending on a half byte boundary, followed by all possible +contexts formed by appending up to 3 more bits. + +A hash table accepts a 32 bit context, which must be a hash if +longer than 4 bytes. The input is further hashed and divided into +an index (depending on the table size, a power of 2), and an 8 bit +checksum which is stored in the table and used to detect collisions +(not perfectly). A lookup tests 3 adjacent locations within a single +64 byte cache line, and if a matching checksum is not found, then the +entry with the smallest value in the first data element is replaced +(LFU replacement policy). This element represents a bit history +for a context ending on a half byte boundary. The states are ordered +so that larger values represent larger total bit counts, which +estimates the likelihood of future use. The initial state is 0. + +Memory is allocated from MEM = pow(2, opt+22) bytes, where opt is 1 through +9 (user selected). Of this, MEM/2 is for the hash table for storing literal +context states, MEM/8 for the rotating LZP buffer, and MEM/8 for a +hash table of pointers into the buffer, plus 12 MB for miscellaneous data. +Total memory usage is 0.75*MEM + 12 MB. + + +ARITHMETIC CODING + +The arithmetic coder codes a bit with probability p using log2(1/p) bits. +Given input string y, the output is a binary fraction x such that +P(< y) <= x < P(<= y) where P(< y) means the total probability of all inputs +lexicographically less than y and P(<= y) = P(< y) + P(y). Note that one +can always find x with length at most log2(P(y)) + 1 bits. + +x can be computed efficiently by maintaining a range, low <= x < high +(initially 0..1) and expressing P(y) as a product of predictions: +P(y) = P(y1) P(y2|y1) P(y3|y1y2) P(y4|y1y2y3) ... P(yn|y1y2...yn-1) +where the term P(yi|y0y1...yi-1) means the probability that yi is 1 +given the context y1...yi-1, the previous i-1 bits of y. For each +prediction p, the range is split in proportion to the probabilities +of 0 and 1, then updated by taking the half corresponding to the actual +bit y as the new range, i.e. + + mid = low + (high - low) * p(y = 1) + if y = 0 then (low, high) := (mid, high) + if y = 1 then (low, high) := (low, mid) + +As low and high approach each other, the high order bits of x become +known (because they are the same throughout the range) and can be +output immediately. + +For decoding, the range is split as before and the range is updated +to the half containing x. The corresponding bit y is used to update +the model. Thus, the model has the same knowledge for coding and +decoding. + +*/ + +#include +#include +#include +#include +#include +#define NDEBUG // remove for debugging +#include + +int allocated=0; // Total memory allocated by alloc() + +// Create an array p of n elements of type T +template void alloc(T*&p, int n) { + p=(T*)calloc(n, sizeof(T)); + if (!p) printf("Out of memory\n"), exit(1); + allocated+=n*sizeof(T); +} + +// 8, 16, 32 bit unsigned types (adjust as appropriate) +typedef unsigned char U8; +typedef unsigned short U16; +typedef unsigned int U32; + +///////////////////////////// Squash ////////////////////////////// + +// return p = 1/(1 + exp(-d)), d scaled by 8 bits, p scaled by 12 bits +class Squash { + short tab[4096]; +public: + Squash(); + int operator()(int d) { + d+=2048; + if (d<0) return 0; + else if (d>4095) return 4095; + else return tab[d]; + } +} squash; + +Squash::Squash() { + static const int t[33]={ + 1,2,3,6,10,16,27,45,73,120,194,310,488,747,1101, + 1546,2047,2549,2994,3348,3607,3785,3901,3975,4022, + 4050,4068,4079,4085,4089,4092,4093,4094}; + for (int i=-2048; i<2048; ++i) { + int w=i&127; + int d=(i>>7)+16; + tab[i+2048]=(t[d]*(128-w)+t[(d+1)]*w+64) >> 7; + } +} + +//////////////////////////// Stretch /////////////////////////////// + +// Inverse of squash. stretch(d) returns ln(p/(1-p)), d scaled by 8 bits, +// p by 12 bits. d has range -2047 to 2047 representing -8 to 8. +// p has range 0 to 4095 representing 0 to 1. + +class Stretch { + short t[4096]; +public: + Stretch(); + int operator()(int p) const { + assert(p>=0 && p<4096); + return t[p]; + } +} stretch; + +Stretch::Stretch() { + int pi=0; + for (int x=-2047; x<=2047; ++x) { // invert squash() + int i=squash(x); + for (int j=pi; j<=i; ++j) + t[j]=x; + pi=i+1; + } + t[4095]=2047; +} + +///////////////////////////// ilog ////////////////////////////// + +// ilog(x) = round(log2(x) * 16), 0 <= x < 64K +class Ilog { + U8* t; +public: + int operator()(U16 x) const {return t[x];} + Ilog(); +} ilog; + +// Compute lookup table by numerical integration of 1/x +Ilog::Ilog() { + alloc(t, 65536); + U32 x=14155776; + for (int i=2; i<65536; ++i) { + x+=774541002/(i*2-1); // numerator is 2^29/ln 2 + t[i]=x>>24; + } +} + +// llog(x) accepts 32 bits +inline int llog(U32 x) { + if (x>=0x1000000) + return 256+ilog(x>>16); + else if (x>=0x10000) + return 128+ilog(x>>8); + else + return ilog(x); +} + +///////////////////////// state table //////////////////////// + +// State table: +// nex(state, 0) = next state if bit y is 0, 0 <= state < 256 +// nex(state, 1) = next state if bit y is 1 +// +// States represent a bit history within some context. +// State 0 is the starting state (no bits seen). +// States 1-30 represent all possible sequences of 1-4 bits. +// States 31-252 represent a pair of counts, (n0,n1), the number +// of 0 and 1 bits respectively. If n0+n1 < 16 then there are +// two states for each pair, depending on if a 0 or 1 was the last +// bit seen. +// If n0 and n1 are too large, then there is no state to represent this +// pair, so another state with about the same ratio of n0/n1 is substituted. +// Also, when a bit is observed and the count of the opposite bit is large, +// then part of this count is discarded to favor newer data over old. + +static const U8 State_table[256][2]={ +{ 1, 2},{ 3, 5},{ 4, 6},{ 7, 10},{ 8, 12},{ 9, 13},{ 11, 14}, // 0 +{ 15, 19},{ 16, 23},{ 17, 24},{ 18, 25},{ 20, 27},{ 21, 28},{ 22, 29}, // 7 +{ 26, 30},{ 31, 33},{ 32, 35},{ 32, 35},{ 32, 35},{ 32, 35},{ 34, 37}, // 14 +{ 34, 37},{ 34, 37},{ 34, 37},{ 34, 37},{ 34, 37},{ 36, 39},{ 36, 39}, // 21 +{ 36, 39},{ 36, 39},{ 38, 40},{ 41, 43},{ 42, 45},{ 42, 45},{ 44, 47}, // 28 +{ 44, 47},{ 46, 49},{ 46, 49},{ 48, 51},{ 48, 51},{ 50, 52},{ 53, 43}, // 35 +{ 54, 57},{ 54, 57},{ 56, 59},{ 56, 59},{ 58, 61},{ 58, 61},{ 60, 63}, // 42 +{ 60, 63},{ 62, 65},{ 62, 65},{ 50, 66},{ 67, 55},{ 68, 57},{ 68, 57}, // 49 +{ 70, 73},{ 70, 73},{ 72, 75},{ 72, 75},{ 74, 77},{ 74, 77},{ 76, 79}, // 56 +{ 76, 79},{ 62, 81},{ 62, 81},{ 64, 82},{ 83, 69},{ 84, 71},{ 84, 71}, // 63 +{ 86, 73},{ 86, 73},{ 44, 59},{ 44, 59},{ 58, 61},{ 58, 61},{ 60, 49}, // 70 +{ 60, 49},{ 76, 89},{ 76, 89},{ 78, 91},{ 78, 91},{ 80, 92},{ 93, 69}, // 77 +{ 94, 87},{ 94, 87},{ 96, 45},{ 96, 45},{ 48, 99},{ 48, 99},{ 88,101}, // 84 +{ 88,101},{ 80,102},{103, 69},{104, 87},{104, 87},{106, 57},{106, 57}, // 91 +{ 62,109},{ 62,109},{ 88,111},{ 88,111},{ 80,112},{113, 85},{114, 87}, // 98 +{114, 87},{116, 57},{116, 57},{ 62,119},{ 62,119},{ 88,121},{ 88,121}, // 105 +{ 90,122},{123, 85},{124, 97},{124, 97},{126, 57},{126, 57},{ 62,129}, // 112 +{ 62,129},{ 98,131},{ 98,131},{ 90,132},{133, 85},{134, 97},{134, 97}, // 119 +{136, 57},{136, 57},{ 62,139},{ 62,139},{ 98,141},{ 98,141},{ 90,142}, // 126 +{143, 95},{144, 97},{144, 97},{ 68, 57},{ 68, 57},{ 62, 81},{ 62, 81}, // 133 +{ 98,147},{ 98,147},{100,148},{149, 95},{150,107},{150,107},{108,151}, // 140 +{108,151},{100,152},{153, 95},{154,107},{108,155},{100,156},{157, 95}, // 147 +{158,107},{108,159},{100,160},{161,105},{162,107},{108,163},{110,164}, // 154 +{165,105},{166,117},{118,167},{110,168},{169,105},{170,117},{118,171}, // 161 +{110,172},{173,105},{174,117},{118,175},{110,176},{177,105},{178,117}, // 168 +{118,179},{110,180},{181,115},{182,117},{118,183},{120,184},{185,115}, // 175 +{186,127},{128,187},{120,188},{189,115},{190,127},{128,191},{120,192}, // 182 +{193,115},{194,127},{128,195},{120,196},{197,115},{198,127},{128,199}, // 189 +{120,200},{201,115},{202,127},{128,203},{120,204},{205,115},{206,127}, // 196 +{128,207},{120,208},{209,125},{210,127},{128,211},{130,212},{213,125}, // 203 +{214,137},{138,215},{130,216},{217,125},{218,137},{138,219},{130,220}, // 210 +{221,125},{222,137},{138,223},{130,224},{225,125},{226,137},{138,227}, // 217 +{130,228},{229,125},{230,137},{138,231},{130,232},{233,125},{234,137}, // 224 +{138,235},{130,236},{237,125},{238,137},{138,239},{130,240},{241,125}, // 231 +{242,137},{138,243},{130,244},{245,135},{246,137},{138,247},{140,248}, // 238 +{249,135},{250, 69},{ 80,251},{140,252},{249,135},{250, 69},{ 80,251}, // 245 +{140,252},{ 0, 0},{ 0, 0},{ 0, 0}}; // 252 +#define nex(state,sel) State_table[state][sel] + +//////////////////////////// StateMap ////////////////////////// + +// A StateMap maps a context to a probability. Methods: +// +// Statemap sm(n) creates a StateMap with n contexts using 4*n bytes memory. +// sm.p(cx, limit) converts state cx (0..n-1) to a probability (0..4095) +// that the next updated bit y=1. +// limit (1..1023, default 255) is the maximum count for computing a +// prediction. Larger values are better for stationary sources. +// sm.update(y) updates the model with actual bit y (0..1). + +class StateMap { +protected: + const int N; // Number of contexts + int cxt; // Context of last prediction + U32 *t; // cxt -> prediction in high 22 bits, count in low 10 bits + static int dt[1024]; // i -> 16K/(i+3) +public: + StateMap(int n=256); + + // update bit y (0..1) + void update(int y, int limit=255) { + assert(cxt>=0 && cxt>10; // count, prediction + if (n>3)*dt[n]&0xfffffc00; + } + + // predict next bit in context cx + int p(int cx) { + assert(cx>=0 && cx>20; + } +}; + +int StateMap::dt[1024]={0}; + +StateMap::StateMap(int n): N(n), cxt(0) { + alloc(t, N); + for (int i=0; i=0 && cx>16)+(x2=p2)*(wt[cxt+1]>>16)+128>>8; + } + void update(int y) { + assert(y==0 || y==1); + int err=((y<<12)-squash(pr)); + if ((wt[cxt]&3)<3) + err*=4-(++wt[cxt]&3); + err=err+8>>4; + wt[cxt]+=x1*err&-4; + wt[cxt+1]+=x2*err; + } +}; + +Mix::Mix(int n): N(n), x1(0), x2(0), cxt(0), pr(0) { + alloc(wt, n*2); + for (int i=0; i h(n) - create using n bytes n and B must be +// powers of 2 with n >= B*4, and B >= 2. +// h[i] returns array [1..B-1] of bytes indexed by i, creating and +// replacing another element if needed. Element 0 is the +// checksum and should not be modified. + +template +class HashTable { + U8* t; // table: 1 element = B bytes: checksum priority data data + const U32 N; // size in bytes +public: + HashTable(int n); + ~HashTable(); + U8* operator[](U32 i); +}; + +template +HashTable::HashTable(int n): t(0), N(n) { + assert(B>=2 && (B&B-1)==0); + assert(N>=B*4 && (N&N-1)==0); + alloc(t, N+B*4+64); + t+=64-int(((long)t)&63); // align on cache line boundary +} + +template +inline U8* HashTable::operator[](U32 i) { + i*=123456791; + i=i<<16|i>>16; + i*=234567891; + int chk=i>>24; + i=i*B&N-B; + if (t[i]==chk) return t+i; + if (t[i^B]==chk) return t+(i^B); + if (t[i^B*2]==chk) return t+(i^B*2); + if (t[i+1]>t[i+1^B] || t[i+1]>t[i+1^B*2]) i^=B; + if (t[i+1]>t[i+1^B^B*2]) i^=B^B*2; + memset(t+i, 0, B); + t[i]=chk; + return t+i; +} + +template +HashTable::~HashTable() { + int c=0, c0=0; + for (U32 i=0; i %1.4f%% full, %1.4f%% utilized of %d KiB\n", + B, 100.0*c0*B/N, 100.0*c/N, N>>10); +} + +////////////////////////// LZP ///////////////////////// + +U32 MEM=1<<29; // Global memory limit, 1 << 22+(memory option) + +// LZP predicts the next byte and maintains context. Methods: +// c() returns the predicted byte for the next update, or -1 if none. +// p() returns the 12 bit probability (0..4095) that c() is next. +// update(ch) updates the model with actual byte ch (0..255). +// c(i) returns the i'th prior byte of context, i > 0. +// c4() returns the order 4 context, shifted into the LSB. +// c8() returns a hash of the order 8 context, shifted 4 bits into LSB. +// word0, word1 are hashes of the current and previous word (a-z). + +class LZP { +private: + const int N, H; // buf, t sizes + enum {MINLEN=12}; // minimum match length + U8* buf; // Rotating buffer of size N + U32* t; // hash table of pointers in high 24 bits, state in low 8 bits + int match; // start of match + int len; // length of match + int pos; // position of next ch to write to buf + U32 h; // context hash + U32 h1; // hash of last 8 byte updates, shifting 4 bits to MSB + U32 h2; // last 4 updates, shifting 8 bits to MSB + StateMap sm1; // len+offset -> p + APM a1, a2, a3; // p, context -> p + int literals, matches; // statistics +public: + U32 word0, word1; // hashes of last 2 words (case insensitive a-z) + LZP(); + ~LZP(); + int c(); // predicted char + int c(int i);// context + int c4() {return h2;} // order 4 context, c(1) in LSB + int c8() {return h1;} // hashed order 8 context + int p(); // probability that next char is c() * 4096 + void update(int ch); // update model with actual char ch +}; + +// Initialize +LZP::LZP(): N(MEM/8), H(MEM/32), + match(-1), len(0), pos(0), h(0), h1(0), h2(0), + sm1(0x200), a1(0x10000), a2(0x40000), a3(0x100000), + literals(0), matches(0), word0(0), word1(0) { + assert(MEM>0); + assert(H>0); + alloc(buf, N); + alloc(t, H); +} + +// Print statistics +LZP::~LZP() { + int c=0; + for (int i=0; i>8, pos>10); + printf("LZP %d literals, %d matches (%1.4f%% matched)\n", + literals, matches, + literals+matches>0?100.0*matches/(literals+matches):0.0); +} + +// Predicted next byte, or -1 for no prediction +inline int LZP::c() { + return len>=MINLEN ? buf[match&N-1] : -1; +} + +// Return i'th byte of context (i > 0) +inline int LZP::c(int i) { + assert(i>0); + return buf[pos-i&N-1]; +} + +// Return prediction that c() will be the next byte (0..4095) +int LZP::p() { + if (len28) cxt=28+(len>=32)+(len>=64)+(len>=128); + int pc=c(); + int pr=sm1.p(cxt); + pr=stretch(pr); + pr=a1.pp(2048, pr*2, h2*256+pc&0xffff)*3+pr>>2; + pr=a2.pp(2048, pr*2, h1*(11<<6)+pc&0x3ffff)*3+pr>>2; + pr=a3.pp(2048, pr*2, h1*(7<<4)+pc&0xfffff)*3+pr>>2; + pr=squash(pr); + return pr; +} + +// Update model with predicted byte ch (0..255) +void LZP::update(int ch) { + int y=c()==ch; // 1 if prediction of ch was right, else 0 + h1=h1*(3<<4)+ch+1; // update context hashes + h2=h2<<8|ch; + h=h*(5<<2)+ch+1&H-1; + if (len>=MINLEN) { + sm1.update(y); + a1.update(y); + a2.update(y); + a3.update(y); + } + if (isalpha(ch)) + word0=word0*(29<<2)+tolower(ch); + else if (word0) + word1=word0, word0=0; + buf[pos&N-1]=ch; // update buf + ++pos; + if (y) { // extend match + ++len; + ++match; + ++matches; + } + else { // find new match, try order 6 context first + ++literals; + y=0; + len=1; + match=t[h]; + if (!((match^pos)&N-1)) --match; + while (len<=128 && buf[match-len&N-1]==buf[pos-len&N-1]) ++len; + --len; + } + t[h]=pos; +} + +LZP* lzp=0; + +//////////////////////////// Predictor ///////////////////////// + +// A Predictor estimates the probability that the next bit of +// uncompressed data is 1. Methods: +// Predictor() creates. +// p() returns P(1) as a 12 bit number (0-4095). +// update(y) trains the predictor with the actual bit (0 or 1). + +class Predictor { + enum {N=11}; // number of contexts + int c0; // last 0-7 bits with leading 1, 0 before LZP flag + int nibble; // last 0-3 bits with leading 1 (1..15) + int bcount; // number of bits in c0 (0..7) + HashTable<16> t; // context -> state + StateMap sm[N]; // state -> prediction + U8* cp[N]; // i -> state array of bit histories for i'th context + U8* sp[N]; // i -> pointer to bit history for i'th context + Mix m[N-1]; // combines 2 predictions given a context + APM a1, a2, a3; // adjusts a prediction given a context + U8* t2; // order 1 contexts -> state + +public: + Predictor(); + int p(); + void update(int y); +}; + +// Initialize +Predictor::Predictor(): + c0(0), nibble(1), bcount(0), t(MEM/2), + a1(0x10000), a2(0x10000), a3(0x10000) { + alloc(t2, 0x40000); + for (int i=0; i=0 && bcount<8); + assert(c0>=0 && c0<256); + assert(nibble>=1 && nibble<=15); + if (c0==0) + c0=1-y; + else { + *sp[0]=nex(*sp[0], y); + sm[0].update(y); + for (int i=1; i=16) nibble=1; + a1.update(y); + a2.update(y); + a3.update(y); + } +} + +// Predict next bit +int Predictor::p() { + assert(lzp); + if (c0==0) + return lzp->p(); + else { + + // Set context pointers + int pc=lzp->c(); // mispredicted byte + int r=pc+256>>8-bcount==c0; // c0 consistent with mispredicted byte? + U32 c4=lzp->c4(); // last 4 whole context bytes, shifted into LSB + U32 c8=(lzp->c8()<<4)-1; // hash of last 7 bytes with 4 trailing 1 bits + if ((bcount&3)==0) { // nibble boundary? Update context pointers + pc&=-r; + U32 c4p=c4<<8; + if (bcount==0) { // byte boundary? Update order-1 context pointers + cp[0]=t2+(c4>>16&0xff00); + cp[1]=t2+(c4>>8 &0xff00)+0x10000; + cp[2]=t2+(c4 &0xff00)+0x20000; + cp[3]=t2+(c4<<8 &0xff00)+0x30000; + } + cp[4]=t[(c4p&0xffff00)-c0]; + cp[5]=t[(c4p&0xffffff00)*3+c0]; + cp[6]=t[c4*7+c0]; + cp[7]=t[(c8*5&0xfffffc)+c0]; + cp[8]=t[(c8*11&0xffffff0)+c0+pc*13]; + cp[9]=t[lzp->word0*5+c0+pc*17]; + cp[10]=t[lzp->word1*7+lzp->word0*11+c0+pc*37]; + } + + // Mix predictions + r<<=8; + sp[0]=&cp[0][c0]; + int pr=stretch(sm[0].p(*sp[0])); + for (int i=1; i>2; + } + pr=a1.pp(512, pr*2, c0+pc*256&0xffff)*3+pr>>2; // Adjust prediction + pr=a2.pp(512, pr*2, c4<<8&0xff00|c0)*3+pr>>2; + pr=a3.pp(512, pr*2, c4*3+c0&0xffff)*3+pr>>2; + return squash(pr); + } +} + +Predictor* predictor=0; + +/////////////////////////// get4, put4 ////////////////////////// + +// Read/write a 4 byte big-endian number +int get4(FILE* in) { + int r=getc(in); + r=r*256+getc(in); + r=r*256+getc(in); + r=r*256+getc(in); + return r; +} + +void put4(U32 c, FILE* out) { + fprintf(out, "%c%c%c%c", c>>24, c>>16, c>>8, c); +} + +//////////////////////////// Encoder //////////////////////////// + +// An Encoder arithmetic codes in blocks of size BUFSIZE. Methods: +// Encoder(COMPRESS, f) creates encoder for compression to archive f, which +// must be open past any header for writing in binary mode. +// Encoder(DECOMPRESS, f) creates encoder for decompression from archive f, +// which must be open past any header for reading in binary mode. +// code(i) in COMPRESS mode compresses bit i (0 or 1) to file f. +// code() in DECOMPRESS mode returns the next decompressed bit from file f. +// count() should be called after each byte is compressed. +// flush() should be called after compression is done. It is also called +// automatically when a block is written. + +typedef enum {COMPRESS, DECOMPRESS} Mode; +class Encoder { +private: + const Mode mode; // Compress or decompress? + FILE* archive; // Compressed data file + U32 x1, x2; // Range, initially [0, 1), scaled by 2^32 + U32 x; // Decompress mode: last 4 input bytes of archive + enum {BUFSIZE=0x20000}; + static unsigned char* buf; // Compression output buffer, size BUFSIZE + int usize, csize; // Buffered uncompressed and compressed sizes + double usum, csum; // Total of usize, csize + +public: + Encoder(Mode m, FILE* f); + void flush(); // call this when compression is finished + + // Compress bit y or return decompressed bit + int code(int y=0) { + assert(predictor); + int p=predictor->p(); + assert(p>=0 && p<4096); + p+=p<2048; + U32 xmid=x1 + (x2-x1>>12)*p + ((x2-x1&0xfff)*p>>12); + assert(xmid>=x1 && xmidupdate(y); + while (((x1^x2)&0xff000000)==0) { // pass equal leading bytes of range + if (mode==COMPRESS) buf[csize++]=x2>>24; + x1<<=8; + x2=(x2<<8)+255; + if (mode==DECOMPRESS) x=(x<<8)+getc(archive); + } + return y; + } + + // Count one byte + void count() { + assert(mode==COMPRESS); + ++usize; + if (csize>BUFSIZE-256) + flush(); + } +}; +unsigned char* Encoder::buf=0; + +// Create in mode m (COMPRESS or DECOMPRESS) with f opened as the archive. +Encoder::Encoder(Mode m, FILE* f): + mode(m), archive(f), x1(0), x2(0xffffffff), x(0), + usize(0), csize(0), usum(0), csum(0) { + if (mode==DECOMPRESS) { // x = first 4 bytes of archive + for (int i=0; i<4; ++i) + x=(x<<8)+(getc(archive)&255); + csize=4; + } + else if (!buf) + alloc(buf, BUFSIZE); +} + +// Write a compressed block and reinitialize the encoder. The format is: +// uncompressed size (usize, 4 byte, MSB first) +// compressed size (csize, 4 bytes, MSB first) +// compressed data (csize bytes) +void Encoder::flush() { + if (mode==COMPRESS) { + buf[csize++]=x1>>24; + buf[csize++]=255; + buf[csize++]=255; + buf[csize++]=255; + putc(0, archive); + putc('c', archive); + put4(usize, archive); + put4(csize, archive); + fwrite(buf, 1, csize, archive); + usum+=usize; + csum+=csize+10; + printf("%15.0f -> %15.0f" + "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b", + usum, csum); + x1=x=usize=csize=0; + x2=0xffffffff; + } +} + +/////////////////////////// paq9a //////////////////////////////// + +// Compress or decompress from in to out, depending on whether mode +// is COMPRESS or DECOMPRESS. A byte c is encoded as a 1 bit if it +// is predicted by LZP, otherwise a 0 followed by 8 bits from MSB to LSB. +void paq9a(FILE* in, FILE* out, Mode mode) { + if (!lzp && !predictor) { + lzp=new LZP; + predictor=new Predictor; + printf("%8d KiB\b\b\b\b\b\b\b\b\b\b\b\b", allocated>>10); + } + if (mode==COMPRESS) { + Encoder e(COMPRESS, out); + int c; + while ((c=getc(in))!=EOF) { + int cp=lzp->c(); + if (c==cp) + e.code(1); + else + for (int i=8; i>=0; --i) + e.code(c>>i&1); + e.count(); + lzp->update(c); + } + e.flush(); + } + else { // DECOMPRESS + int usize=get4(in); + get4(in); // csize + Encoder e(DECOMPRESS, in); + while (usize--) { + int c=lzp->c(); + if (e.code()==0) { + c=1; + while (c<256) c+=c+e.code(); + c&=255; + } + if (out) putc(c, out); + lzp->update(c); + } + } +} + + +///////////////////////////// store /////////////////////////// + +// Store a file in blocks as: {'\0' mode usize csize contents}... +void store(FILE* in, FILE* out) { + assert(in); + assert(out); + + // Store in blocks + const int BLOCKSIZE=0x100000; + static char* buf=0; + if (!buf) alloc(buf, BLOCKSIZE); + bool first=true; + while (true) { + int n=fread(buf, 1, BLOCKSIZE, in); + if (!first && n<=0) break; + fprintf(out, "%c%c", 0, 's'); + put4(n, out); // usize + put4(n, out); // csize + fwrite(buf, 1, n, out); + first=false; + } + + // Close file + fclose(in); +} + +// Write usize == csize bytes of an uncompressed block from in to out +void unstore(FILE* in, FILE* out) { + assert(in); + int usize=get4(in); + int csize=get4(in); + if (usize!=csize) + printf("Bad archive format: usize=%d csize=%d\n", usize, csize); + static char* buf=0; + const int BUFSIZE=0x1000; + if (!buf) alloc(buf, BUFSIZE); + while (csize>0) { + usize=csize; + if (usize>BUFSIZE) usize=BUFSIZE; + if (int(fread(buf, 1, usize, in))!=usize) + printf("Unexpected end of archive\n"), exit(1); + if (out) fwrite(buf, 1, usize, out); + csize-=usize; + } +} + +//////////////////////// Archiving functions //////////////////////// + +const int MAXNAMELEN=1023; // max filename length + +// Return true if the first 4 bytes of in are a valid archive +bool check_archive(FILE* in) { + return getc(in)=='p' && getc(in)=='Q' && getc(in)=='9' && getc(in)==1; +} + +// Open archive and check for valid archive header, exit if bad. +// Set MEM to memory option '1' through '9' +FILE* open_archive(const char* filename) { + FILE* in=fopen(filename, "rb"); + if (!in) + printf("Cannot find archive %s\n", filename), exit(1); + if (!check_archive(in) || (MEM=getc(in))<'1' || MEM>'9') { + fclose(in); + printf("%s: Not a paq9a archive\n", filename); + exit(1); + } + return in; +} + +// Compress filename to out. option is 'c' to compress or 's' to store. +void compress(const char* filename, FILE* out, int option) { + + // Open input file + FILE* in=fopen(filename, "rb"); + if (!in) { + printf("File not found: %s\n", filename); + return; + } + fprintf(out, "%s", filename); + printf("%-40s ", filename); + + // Compress depending on option + if (option=='s') + store(in, out); + else if (option=='c') + paq9a(in, out, COMPRESS); + printf("\n"); +} + +// List archive contents +void list(const char* archive) { + double usum=0, csum=0; // uncompressed and compressed size per file + double utotal=0, ctotal=4; // total size in archive + static char filename[MAXNAMELEN+1]; + int mode=0; + + FILE* in=open_archive(archive); + printf("\npaq9a -%c\n", MEM); + while (true) { + + // Get filename, mode + int c=getc(in); + if (c==EOF) break; + if (c) { // start of new file? Print previous file + if (mode) + printf("%10.0f -> %10.0f %c %s\n", usum, csum, mode, filename); + int len=0; + filename[len++]=c; + while ((c=getc(in))!=EOF && c) + if (lenBUFSIZE) + csize-=fread(buf, 1, BUFSIZE, in); + fread(buf, 1, csize, in); + } + printf("%10.0f -> %10.0f %c %s\n", usum, csum, mode, filename); + utotal+=usum; + ctotal+=csum; + printf("%10.0f -> %10.0f total\n", utotal, ctotal); + fclose(in); +} + +// Extract files given command line arguments +// Input format is: [filename {'\0' mode usize csize contents}...]... +void extract(int argc, char** argv) { + assert(argc>2); + assert(argv[1][0]=='x'); + static char filename[MAXNAMELEN+1]; // filename from archive + + // Open archive + FILE* in=open_archive(argv[2]); + MEM=1<<22+MEM-'0'; + + // Extract files + argc-=3; + argv+=3; + FILE* out=0; + while (true) { // for each block + + // Get filename + int c; + for (int i=0;; ++i) { + c=getc(in); + if (c==EOF) break; + if (i0) fn=argv[0], --argc, ++argv; + if (out) fclose(out); + out=fopen(fn, "rb"); + if (out) { + printf("\nCannot overwrite file, skipping: %s ", fn); + fclose(out); + out=0; + } + else { + out=fopen(fn, "wb"); + if (!out) printf("\nCannot create file: %s ", fn); + } + if (out) { + if (fn==filename) printf("\n%s ", filename); + else printf("\n%s -> %s ", filename, fn); + } + } + + // Extract block + int mode=getc(in); + if (mode=='s') + unstore(in, out); + else if (mode=='c') + paq9a(in, out, DECOMPRESS); + else + printf("\nUnsupported compression mode %c %d at %ld\n", + mode, mode, ftell(in)), exit(1); + } + printf("\n"); + if (out) fclose(out); +} + +// Command line is: paq9a {a|x|l} archive [[-option] files...]... +int main(int argc, char** argv) { + clock_t start=clock(); + + // Check command line arguments + if (argc<3 || argv[1][1] || (argv[1][0]!='a' && argv[1][0]!='x' + && argv[1][0]!='l') || (argv[1][0]=='a' && argc<4) || argv[2][0]=='-') + { + printf("paq9a archiver (C) 2007, Matt Mahoney\n" + "Free software under GPL, http://www.gnu.org/copyleft/gpl.html\n" + "\n" + "To create archive: paq9a a archive [-1..-9] [[-s|-c] files...]...\n" + " -1..-9 = use 18 to 1585 MiB memory (default -7 = 408 MiB)\n" + " -s = store, -c = compress (default)\n" + "To extract files: paq9a x archive [files...]\n" + "To list contents: paq9a l archive\n"); + exit(1); + } + + // Create archive + if (argv[1][0]=='a') { + int option = 'c'; // -c or -s + FILE* out=fopen(argv[2], "rb"); + if (out) printf("Cannot overwrite archive %s\n", argv[2]), exit(1); + out=fopen(argv[2], "wb"); + if (!out) printf("Cannot create archive %s\n", argv[2]), exit(1); + fprintf(out, "pQ9%c", 1); + int i=3; + if (argc>3 && argv[3][0]=='-' && argv[3][1]>='1' && argv[3][1]<='9' + && argv[3][2]==0) { + putc(argv[3][1], out); + MEM=1<<22+argv[3][1]-'0'; + ++i; + } + else + putc('7', out); + for (; i %ld in %1.2f sec\n", ftell(out), + double(clock()-start)/CLOCKS_PER_SEC); + } + + // List archive contents + else if (argv[1][0]=='l') + list(argv[2]); + + // Extract from archive + else if (argv[1][0]=='x') { + extract(argc, argv); + printf("%1.2f sec\n", double(clock()-start)/CLOCKS_PER_SEC); + } + + // Report statistics + delete predictor; + delete lzp; + printf("Used %d KiB memory\n", allocated>>10); + return 0; +}