diff --git a/.dockerignore b/.dockerignore new file mode 100644 index 0000000..bb34b59 --- /dev/null +++ b/.dockerignore @@ -0,0 +1,9 @@ +.git +.github +.Rproj.user +.Rhistory +.RData +.Ruserdata +*.Rcheck +*.tar.gz +docs/superpowers diff --git a/.github/workflows/test-fast.yaml b/.github/workflows/test-fast.yaml new file mode 100644 index 0000000..fa70edf --- /dev/null +++ b/.github/workflows/test-fast.yaml @@ -0,0 +1,41 @@ +name: test-fast + +on: + push: + branches: [main, master, develop] + pull_request: + branches: [main, master, develop] + +permissions: + contents: read + +jobs: + test-fast: + runs-on: ubuntu-latest + + steps: + - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2 + + - uses: r-lib/actions/setup-r@6f6e5bc62fba3a704f74e7ad7ef7676c5c6a2590 # v2 + with: + use-public-rspm: true + + - uses: r-lib/actions/setup-r-dependencies@6f6e5bc62fba3a704f74e7ad7ef7676c5c6a2590 # v2 + with: + extra-packages: any::testthat + needs: check + + - name: Install kaefa package for fast tests + run: R CMD INSTALL . + + - name: Run fast productization tests + run: | + Rscript - <<'RSCRIPT' + reporter <- testthat::StopReporter$new() + testthat::test_file("tests/testthat/test-benchmark-manifest.R", + reporter = reporter) + testthat::test_file("tests/testthat/test-shiny-product-surface.R", + reporter = reporter) + testthat::test_file("tests/testthat/test-core-api-contract.R", + reporter = reporter) + RSCRIPT diff --git a/ARCHITECTURE.md b/ARCHITECTURE.md index dd31bce..6ddd063 100644 --- a/ARCHITECTURE.md +++ b/ARCHITECTURE.md @@ -11,6 +11,23 @@ It provides: - optional remote worker initialization (`aefaInit`), - an interactive Shiny UI (`launchAEFA`). +## Productization Boundaries + +The repository remains a monorepo until the product surface has independent +release or deployment needs. Use these boundaries when planning sale-readiness +work: + +- `kaefa-core`: the R/statistical engine boundary around `aefa()`, + `engineAEFA()`, model selection, item-fit evaluation, theta-prior utilities, + and benchmark evidence. +- `kaefa-studio`: the buyer-facing UI boundary around `launchAEFA()` and + `inst/shiny-app/app.R`. +- `kaefa-runner`: the future deployment and execution boundary for container, + hosted, remote, or scheduled analysis workflows. + +Do not introduce a git submodule unless a downstream buyer or deployment model +explicitly requires vendored source integration. + ## Repository Layout - `R/kaefa.R`: public orchestration entry points and exported runtime behavior. diff --git a/DESCRIPTION b/DESCRIPTION index 3ad2079..e7ed53d 100644 --- a/DESCRIPTION +++ b/DESCRIPTION @@ -31,6 +31,7 @@ LazyData: true Repository: CRAN Suggests: covr, testthat, + pkgload, knitr, rmarkdown VignetteBuilder: knitr diff --git a/Dockerfile b/Dockerfile new file mode 100644 index 0000000..eb7582d --- /dev/null +++ b/Dockerfile @@ -0,0 +1,28 @@ +FROM r-base:4.4.2 + +ENV R_REPOS=https://cloud.r-project.org + +WORKDIR /opt/kaefa + +RUN apt-get update \ + && apt-get install -y --no-install-recommends \ + cmake \ + libcurl4-openssl-dev \ + libssl-dev \ + libxml2-dev \ + libuv1-dev \ + pandoc \ + xz-utils \ + && rm -rf /var/lib/apt/lists/* + +COPY DESCRIPTION /opt/kaefa/DESCRIPTION + +RUN Rscript -e 'install.packages("remotes", repos = Sys.getenv("R_REPOS")); remotes::install_deps(dependencies = c("Depends", "Imports", "LinkingTo"), repos = Sys.getenv("R_REPOS"), upgrade = "never")' + +COPY . /opt/kaefa + +RUN R CMD INSTALL . + +EXPOSE 3838 + +CMD ["Rscript", "-e", "kaefa::launchAEFA(host = '0.0.0.0', port = 3838)"] diff --git a/LICENSE b/LICENSE index 69fc4ef..f288702 100644 --- a/LICENSE +++ b/LICENSE @@ -1 +1,674 @@ -NA + 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/R/shiny_helpers.R b/R/shiny_helpers.R new file mode 100644 index 0000000..b4384c9 --- /dev/null +++ b/R/shiny_helpers.R @@ -0,0 +1,89 @@ +.kaefaStudioInvalidColumns <- function(data) { + if (is.null(data)) { + return(character()) + } + if (!is.data.frame(data)) { + stop("Studio input data must be a data.frame.", call. = FALSE) + } + + numeric_columns <- vapply(data, is.numeric, logical(1)) + names(data)[!numeric_columns] +} + +.kaefaStudioInvalidColumnMessage <- function(invalid_columns, max_columns = 8L) { + if (!length(invalid_columns)) { + return(NULL) + } + + invalid_preview <- paste(utils::head(invalid_columns, max_columns), + collapse = ", ") + if (length(invalid_columns) > max_columns) { + invalid_preview <- paste0(invalid_preview, ", ...") + } + + paste0( + "All item columns must be numeric for factor analysis. ", + "Non-numeric columns: ", invalid_preview, "." + ) +} + +.kaefaStudioRunOptions <- function(data, input, + package_version = as.character( + utils::packageVersion("kaefa") + ), + started_at = format( + Sys.time(), + "%Y-%m-%dT%H:%M:%SZ", + tz = "UTC" + )) { + list( + packageVersion = package_version, + startedAt = started_at, + rows = nrow(data), + items = ncol(data), + minFactors = input$minFactors, + maxFactors = input$maxFactors, + rotation = input$rotation, + modelSelection = input$modelSelection, + saveHistory = input$saveHistory + ) +} + +.kaefaStudioReportMetadataLines <- function(run_options, data, input) { + option_value <- function(name, fallback) { + if (!is.null(run_options[[name]])) { + return(run_options[[name]]) + } + fallback + } + + c( + "Run Metadata:", + "-------------", + paste( + "kaefa package version:", + option_value( + "packageVersion", + as.character(utils::packageVersion("kaefa")) + ) + ), + paste("Analysis started:", option_value("startedAt", "not recorded")), + paste( + "Data shape:", + option_value("rows", nrow(data)), + "rows x", + option_value("items", ncol(data)), + "items" + ), + "Selected options:", + paste("- Minimum factors:", + option_value("minFactors", input$minFactors)), + paste("- Maximum factors:", + option_value("maxFactors", input$maxFactors)), + paste("- Rotation:", option_value("rotation", input$rotation)), + paste("- Model selection:", + option_value("modelSelection", input$modelSelection)), + paste("- Save model history:", + option_value("saveHistory", input$saveHistory)) + ) +} diff --git a/TRD.md b/TRD.md index a74ae43..0767c89 100644 --- a/TRD.md +++ b/TRD.md @@ -75,6 +75,7 @@ The package exposes a programmatic API and an optional Shiny interface. manager with regular rotation and least-privilege access. See [Security and Privacy](#security-and-privacy). Example usage with `aefaInit()`: + ```r # vector aligned with RemoteClusters aefaInit( @@ -120,6 +121,26 @@ The package exposes a programmatic API and an optional Shiny interface. - R CMD check on Windows, macOS, and Linux in CI. - Unit tests in `tests/` for core logic and regressions. - Example workflows in README and vignettes for smoke validation. +- Fast PR tests should validate metadata, documentation, and lightweight + behavior without requiring heavy model-fitting runs. +- Release benchmark tests should validate selected datasets from + `inst/benchmarks/manifest.csv` before tagged releases or buyer-facing + diligence packages. +- Private benchmark tests should run outside public CI when datasets are + restricted, with only reviewed aggregate evidence committed back to docs. + +## Productization Boundaries + +- `kaefa-core`: R package APIs, statistical engine, benchmark evidence, and + reproducible result objects. +- `kaefa-studio`: bundled Shiny UI, report export, user-facing validation, and + future hosted product workflow. +- `kaefa-runner`: future container, ShinyProxy, remote execution, monitoring, + and scheduled benchmark operations. + +Keep these as documented boundaries before splitting repositories. Split only +when release cadence, runtime dependencies, or buyer deployment requirements +make a separate package or repository materially simpler. ## Security and Privacy diff --git a/deploy/shinyproxy/application.yml.example b/deploy/shinyproxy/application.yml.example new file mode 100644 index 0000000..e116cb6 --- /dev/null +++ b/deploy/shinyproxy/application.yml.example @@ -0,0 +1,20 @@ +proxy: + title: kaefa Studio + port: 8080 + authentication: simple + users: + - name: analyst + password: ${KAEFA_SHINYPROXY_ANALYST_PASSWORD} + groups: kaefa-users + specs: + - id: kaefa-studio + display-name: kaefa Studio + description: Automated exploratory factor analysis workspace + container-image: kaefa-studio:local + container-network: kaefa-net + access-groups: [kaefa-users] + container-volumes: [] + container-memory-request: 4g + container-memory-limit: 8g + container-cpu-request: 2 + container-cpu-limit: 4 diff --git a/docs/business/2b-krw-commercial-model.md b/docs/business/2b-krw-commercial-model.md new file mode 100644 index 0000000..b4890db --- /dev/null +++ b/docs/business/2b-krw-commercial-model.md @@ -0,0 +1,148 @@ +# 2B KRW Commercial Model + +## Purpose + +This model defines what must become true for `kaefa` to support a 2B KRW +acquisition case, equivalent to a 20억 KRW sale target. It is not a present-day +valuation claim. + +## Valuation Threshold + +| Exit multiple | ARR needed for 2B KRW value | +| --- | ---: | +| 3x ARR | 667M KRW | +| 4x ARR | 500M KRW | +| 5x ARR | 400M KRW | +| Strategic IP sale | Lower ARR possible with stronger evidence | + +The practical target is 400M-700M KRW ARR unless the buyer is acquiring +strategic statistical capability rather than recurring revenue. + +## Source Separation + +Use this model as an operating target, not as a standalone appraisal. + +Public-market assumptions: + +- ARR multiple ranges must be refreshed before buyer use. The working 3x-5x + ARR range is a conservative planning proxy based on public SaaS/software + sources available on 2026-07-02. +- Strategic-IP value is allowed only when the buyer thesis is about statistical + capability, validation evidence, and integration leverage rather than current + recurring revenue. + +Public source anchors reviewed: + + + +| Source | Public signal | Use in this model | +| --- | --- | --- | +| [SaaS Capital Index](https://www.saas-capital.com/the-saas-capital-index/) | Defines its revenue multiple on annualized current run-rate revenue and focuses on public B2B SaaS. | Use run-rate ARR, not trailing service revenue, when comparing `kaefa` to SaaS multiples. | +| [SaaS Capital 2026 bootstrapped benchmarks](https://www.saas-capital.com/blog-posts/benchmarking-metrics-for-bootstrapped-saas-companies/) | Reports median bootstrapped SaaS growth, NRR, and GRR benchmarks for $3M-$20M ARR companies. | Use retention and growth as quality gates before claiming a premium multiple. | +| [Clouded Judgement 2026-02-06](https://cloudedjudgement.substack.com/p/clouded-judgement-2626-software-is) | Reports a lower public cloud/software median NTM revenue multiple in early 2026. | Supports keeping the `kaefa` base-case multiple conservative. | +| [Aventis 2026 SaaS multiples](https://aventis-advisors.com/saas-valuation-multiples/) | Reports a March 2026 public SaaS EV/revenue median near the low-to-mid single digits. | Cross-checks the 3x-5x planning range. | +| [L40 SaaS multiples 2026](https://www.l40.com/insights/saas-multiples) | Describes private SaaS median ARR multiples around 4x-5x in 2026. | Supports using 4x as the central planning case. | + + + +Repository-backed evidence: + +- `docs/diligence/data-room-index.md` lists the current buyer evidence package. +- `docs/diligence/contextualwisdomlab-resource-map.md` separates adjacent + ContextualWisdomLab assets from kaefa-owned evidence. +- `docs/validation/benchmark-protocol.md` and `inst/benchmarks/manifest.csv` + define the first validation evidence path. +- `docs/product/kaefa-studio-requirements.md` defines the current Studio + product scope and split criteria. + +## Pricing Paths + +### Annual Institution License + +Sell annual access to `kaefa-studio`, validation reports, and support for +university labs, assessment vendors, and research teams. + +Evidence needed: + +- signed annual contracts, +- named user or site scope, +- support SLA, +- renewal terms, +- usage logs that do not expose private respondent data. + +### Hosted Assessment Analytics Workspace + +Sell a managed deployment where the buyer or customer uploads datasets and +receives standardized reports. + +Evidence needed: + +- deployment architecture, +- privacy and retention policy, +- runtime cost per analysis, +- uptime and monitoring proof, +- reproducibility bundle export. + +### Productized Validation Services + +Sell expert-assisted validation projects where the repeatable software output is +explicitly separated from consulting labor. + +Evidence needed: + +- standard statement of work, +- fixed report template, +- before/after analyst time saved, +- conversion path from service project to subscription. + +## Minimum Sale-Readiness Metrics + +- 3-5 paid pilots completed with written acceptance. +- 400M-700M KRW ARR path defined from signed annual contracts or hosted + subscription commitments. +- Median time-to-report measured on at least three benchmark classes. +- At least one repeatable deployment path. +- License/IP posture reviewed. +- Fast PR tests and release benchmark tests documented. +- Known failure classes return structured explanations. + +## ARR And Customer Count Targets + +The 20억 KRW sale target is only credible as a revenue-multiple story if the +contract base can support one of these operating cases: + +| Case | Multiple | ARR target | Example contract base | +| --- | ---: | ---: | --- | +| Conservative | 3x ARR | 667M KRW | 34 institutions at 20M KRW/year | +| Central | 4x ARR | 500M KRW | 25 institutions at 20M KRW/year | +| Quality niche | 5x ARR | 400M KRW | 20 institutions at 20M KRW/year | + +If pricing lands closer to 10M KRW/year, the central case needs about 50 +recurring customers. If pricing lands closer to 50M KRW/year through enterprise +or assessment-vendor licenses, the central case needs about 10 recurring +customers. Services revenue may help fund pilots, but it should not be counted +as ARR unless the renewal commitment is explicit. + +## KPI Targets For Sale Readiness + + + +| KPI | Minimum target before buyer claim | Evidence source | +| --- | --- | --- | +| Paid pilots | 3 completed pilots, 5 preferred | Signed scope, acceptance note, pilot scorecard | +| Repeatable ARR | 400M-700M KRW run-rate | Contracts or renewal commitments | +| Gross revenue retention | 90%+ once renewals exist | Renewal log and churn notes | +| Net revenue retention | 103%+ once expansion exists | Renewal plus expansion log | +| Time to report | Median under 30 minutes for pilot-sized datasets | Benchmark and Shiny report metadata | +| Report success rate | 90%+ on accepted pilot datasets | Pilot scorecard and benchmark manifest | +| Failure explainability | 100% structured reason for known failure classes | Tests and exported report metadata | + + + +## Buyer Story + +The strongest buyer story is not "an R package exists." It is: + +`kaefa` turns complex IRT/EFA model search into a repeatable report workflow, +with benchmark evidence, deployable UI, and supportable operations for +assessment teams that cannot afford manual model exploration for every dataset. diff --git a/docs/business/pilot-scorecard.md b/docs/business/pilot-scorecard.md new file mode 100644 index 0000000..b30d94e --- /dev/null +++ b/docs/business/pilot-scorecard.md @@ -0,0 +1,55 @@ +# Pilot Scorecard + +## Purpose + +Every paid or strategic pilot should produce comparable evidence for product +fit, technical reliability, and acquisition readiness. + +## Scorecard Fields + +| Field | Scoring Guidance | +| --- | --- | +| Pilot owner | Named accountable owner | +| Customer segment | University, vendor, HR, education, or consulting | +| Dataset complexity | Rows, items, response type, missingness, covariates | +| Successful report generation | Yes, partial, or no | +| Time to report | Wall-clock time from upload or API call to export | +| Analyst intervention | None, light, heavy, or impossible | +| Failure explainability | Structured reason and next action, or raw error | +| Security constraints | Local, hosted, private runner, or restricted | +| Willingness to pay | Annual license, service fee, support fee, or none | +| Renewal path | Clear, possible, unclear, or no | +| Reference value | Public reference, private reference, or internal only | + +## Acceptance Levels + +### Green + +- Report generated successfully. +- Result was understandable to the pilot owner. +- No private data handling concern remains unresolved. +- Customer expresses a paid renewal or expansion path. + +### Yellow + +- Report generated with analyst intervention. +- Runtime or UX issues are tolerable but must be fixed. +- Customer value exists, but packaging or pricing is unclear. + +### Red + +- Model fails without a structured explanation. +- Dataset cannot be used under the current privacy model. +- Customer does not see a paid path after the pilot. + +## Evidence To Save + +- signed pilot scope or email approval, +- dataset metadata without private respondent data, +- selected options, +- runtime measurement, +- exported report, +- customer feedback, +- renewal or expansion decision. + +Do not commit private datasets, credentials, or customer-identifying raw data. diff --git a/docs/diligence/contextualwisdomlab-resource-map.md b/docs/diligence/contextualwisdomlab-resource-map.md new file mode 100644 index 0000000..e945980 --- /dev/null +++ b/docs/diligence/contextualwisdomlab-resource-map.md @@ -0,0 +1,95 @@ +# ContextualWisdomLab Resource Map + +This map records the live organization resources reviewed while shaping the +`kaefa` 20억 KRW sale-readiness plan. It separates assets that can strengthen a +buyer story from assets that must not be treated as `kaefa` evidence until they +are integrated, licensed, and validated. + +## Live Scan + +- Organization: `ContextualWisdomLab`. +- Scan method: `gh repo list ContextualWisdomLab --no-archived` plus per-repo + `gh pr list`, `gh search prs --owner ContextualWisdomLab`, and + `gh search issues --owner ContextualWisdomLab`. +- Scan time: 2026-07-03 00:04 UTC / 2026-07-03 KST. +- `kaefa` default branch: `develop`. +- `kaefa` open PRs: `#61` sale-readiness roadmap plus Dependabot PRs `#55`, + `#57`, `#58`, `#59`, and `#60`. +- `kaefa` open issues: GPU integration `#49`, `nonnest2` model comparison + `#48`, benchmark calibration `#46`, and broad lint/refactor `#45`. +- Related organization resources include `noema` issue `#5` for 20억 readiness + evidence, open `noema` PR `#6` for data-room manifest evidence, and adjacent + statistical assets with live PR queues in `fast-mlsirm`, `nonnest2`, and + `aFIPC`. + +GitHub review state and queued checks are not product blockers by themselves. +Failed CI, broken tests, security findings, license/IP uncertainty, and missing +runtime evidence remain blockers. + +## Adjacent Assets + +### `ContextualWisdomLab/aFIPC` + +R/C++ IRT equating asset that may support an assessment analytics portfolio +story. Reference it only after API, license, validation, and maintenance posture +are reviewed. + +### `ContextualWisdomLab/fast-mlsirm` + +Fast MLSIRM/MLS2PLM simulation and recovery diagnostics adjacent to +psychometric validation. Treat it as a candidate benchmark or future engine +integration only after method review and reproducible comparison. + +### `ContextualWisdomLab/nonnest2` + +Existing `kaefa` issue `#48` names it for model comparison. Treat it as a +planned validation dependency, not current product functionality. + +### `ContextualWisdomLab/noema` + +Organization PR-review automation and evidence-gate infrastructure. Use it for +buyer-facing engineering governance evidence, not as product runtime. + +### `ContextualWisdomLab/.github` + +Shared workflows, review gates, and organization profile assets. Use it as the +source for CI/governance posture; the open PR queue there may affect automation +reliability. + +### `ContextualWisdomLab/appguardrail` + +Security guardrail product that may become a diligence checklist input. Use it +as organizational security capability evidence only if its own release posture +is clean. + +### `ContextualWisdomLab/bandscope`, `pg-erd-cloud`, `naruon` + +Other productization efforts with active PR queues and diligence patterns. Use +them as pattern references only; do not mix their product metrics into `kaefa` +valuation. + +## Open PR Queue Signal + +The organization has a large live PR queue across product, security, UX, and +automation repositories. That is useful as evidence of active engineering +throughput, but it is also a diligence risk unless each buyer-facing repository +has clear release ownership and green required checks. + +For `kaefa`, the immediate action is narrower: + +1. Land PR `#61` once required checks pass or auto-merge can carry queued + checks through the normal rules. +2. Keep Dependabot PRs separate unless they unblock productization evidence. +3. Use adjacent repositories only as optional portfolio/context evidence. +4. Do not create submodules to adjacent repositories before a concrete runtime + contract exists. + +## Figma/FigJam Resources + +- Primary roadmap board: . +- Secondary scratch board created for continued planning: + . +- Figma Code Connect was not used. +- The primary board now includes a `kaefa 2B KRW Productization Architecture` + flow that connects repository boundaries, validation gates, buyer evidence, + adjacent organization assets, and remaining sale-claim blockers. diff --git a/docs/diligence/data-room-index.md b/docs/diligence/data-room-index.md new file mode 100644 index 0000000..1fa7202 --- /dev/null +++ b/docs/diligence/data-room-index.md @@ -0,0 +1,110 @@ +# Data Room Index + +Use this index when packaging `kaefa` for a buyer, pilot sponsor, or strategic +partner. It points to the current evidence and separates completed artifacts +from decisions that still need owner, legal, or customer input. + +## Product And Architecture + +- Product scope: + `docs/product/kaefa-studio-requirements.md`. + Status: ready for pilot review. +- Core API contract: + `docs/product/kaefa-core-api-contract.md`. + Status: pilot-facing API documented; separate package split still deferred. +- Internal boundaries: + `ARCHITECTURE.md`, `TRD.md`. + Status: monorepo boundaries documented. +- Repository split decision: + `docs/diligence/repository-boundary-decision.md`, + `docs/product/kaefa-studio-requirements.md`. + Status: keep one repository with explicit `kaefa-core`, `kaefa-studio`, and + `kaefa-runner` boundaries until split triggers are met. + +## Commercial Evidence + +- Acquisition case: + `docs/business/2b-krw-commercial-model.md`. + Status: target model documented; public-market assumptions are separated + from repository-backed evidence. +- Organization resource map: + `docs/diligence/contextualwisdomlab-resource-map.md`. + Status: live ContextualWisdomLab repository and PR resources reviewed. +- Pilot scoring: + `docs/business/pilot-scorecard.md`. + Status: scorecard ready. +- Pilot evidence template: + `docs/diligence/pilot-evidence-template.md`. + Status: ready for first paid or strategic pilot. +- Missing revenue proof: + `docs/business/2b-krw-commercial-model.md`. + Status: paid pilots and ARR evidence needed. + +## Validation And Quality + +- Fast PR tests: + `.github/workflows/test-fast.yaml`, `tests/FAST_TESTS.md`. + Status: fast productization gate added. +- Benchmark evidence: + `inst/benchmarks/manifest.csv`, + `docs/validation/benchmark-protocol.md`. + Status: manifest and protocol ready. +- Release checks: + `docs/diligence/release-diligence-checklist.md`. + Status: checklist ready. +- Release evidence log: + `docs/diligence/release-evidence-log.md`. + Status: ready for buyer-facing release evidence capture. + +## Deployment And Operations + +- Studio runtime: + `Dockerfile`, `docs/operations/deployment.md`. + Status: runtime-code container build and local Shiny smoke captured on + 2026-07-03 KST for commit `b5bfdb8b509eb8ec06f143c6435f880eda3d2e20`. + `curl http://127.0.0.1:3838/` returned HTTP 200, 12,017 bytes, and title + `kaefa: Automated Exploratory Factor Analysis`. Rerun for release-candidate + signoff and after runtime, Dockerfile, or package dependency changes. +- Hosted deployment example: + `deploy/shinyproxy/application.yml.example`. + Status: evaluation example ready. +- Production hardening: + `docs/operations/deployment.md`. + Status: auth, HTTPS, secrets, resource sizing, and measured runtime budgets + remain. + +## Security Privacy And IP + +- Uploaded data handling: + `docs/product/kaefa-studio-requirements.md`, + `docs/operations/deployment.md`. + Status: no persistence unless user exports. +- License posture: + `LICENSE`, `DESCRIPTION`, `docs/diligence/license-and-ip.md`. + Status: GPL posture documented. +- Legal decisions: + `docs/diligence/license-and-ip.md`. + Status: owner/legal confirmation required before proprietary claims. + +## Design And Roadmap + +- Productization roadmap: + `docs/superpowers/plans/2026-07-02-kaefa-2b-krw-sale-readiness.md`. + Status: execution checklist current. +- Visual roadmap: + FigJam roadmap: . + Status: created and extended with the productization architecture flow + without Figma Code Connect. +- Planning scratch board: + FigJam board: . + Status: created without Figma Code Connect; use as a working board if the + primary roadmap board needs a clean follow-up surface. + +## Missing Before Sale Claim + +- Owner/legal signoff on copyright, GPL dependency implications, and buyer + deliverables. +- At least 3-5 paid or strategic pilots scored with the pilot scorecard. +- Benchmark runtime and accuracy evidence for agreed dataset classes. +- Production authentication, HTTPS, secret management, monitoring, and support + posture if the sale includes hosted operation. diff --git a/docs/diligence/license-and-ip.md b/docs/diligence/license-and-ip.md new file mode 100644 index 0000000..8e9cb5b --- /dev/null +++ b/docs/diligence/license-and-ip.md @@ -0,0 +1,65 @@ +# License And IP Diligence + +## Current Package License + +`DESCRIPTION` currently declares: + +```text +License: GPL-3 + file LICENSE +``` + +The repository `LICENSE` file now begins: + +```text +GNU GENERAL PUBLIC LICENSE +Version 3, 29 June 2007 +``` + +The prior placeholder `LICENSE` file has been replaced with the GPL-3 text +distributed with the local R runtime. This resolves the repository-file +mismatch, but it does not resolve ownership, contributor-rights, or commercial +redistribution decisions. + +## Dependency Constraints + +The containerized Studio runtime confirmed these dependency license facts: + +- `mirt`: GPL (>= 3) +- `future`: LGPL (>= 2.1) +- `shiny`: MIT + file LICENSE +- `DT`: MIT + file LICENSE +- `fitdistrplus`: GPL (>= 2) + +## Working Commercial Posture + +Until owner and legal review explicitly approve another structure, treat the +sellable product as: + +- GPL-compatible statistical core, +- paid implementation and support services, +- optional hosted deployment operated by the seller or buyer, +- commercial reporting templates and benchmark evidence packaged around the + core. + +Do not claim a proprietary source-code sale, dual license, or open-core split +until copyright ownership and GPL dependency implications are reviewed. + +## Required Owner And Legal Checks + +- Confirm copyright ownership for all substantial contributions. +- Confirm whether every contributor assigned rights or contributed under terms + compatible with the intended sale. +- Confirm whether GPL dependencies constrain proprietary redistribution or + embedding. +- Decide whether the buyer receives source code, hosted service rights, + support/maintenance rights, or a mixed package. +- Confirm whether `License: GPL-3 + file LICENSE` should remain as-is or be + simplified for CRAN/package convention while preserving the repository GPL-3 + notice. + +## Recommended Next Decision + +Keep the current GPL-compatible R package as `kaefa-core` unless legal review +authorizes a dual-license model. Build sale value through benchmark evidence, +enterprise deployment, support, reporting, and paid pilots before attempting a +repository split. diff --git a/docs/diligence/pilot-evidence-template.md b/docs/diligence/pilot-evidence-template.md new file mode 100644 index 0000000..dc0de90 --- /dev/null +++ b/docs/diligence/pilot-evidence-template.md @@ -0,0 +1,76 @@ +# Pilot Evidence Template + +Use one copy of this template for each paid or strategic pilot. Keep private +datasets, raw respondent records, credentials, and customer-identifying data out +of git. Commit only metadata, aggregate outcomes, and approved quotes. + +## Pilot Identity + +- Pilot ID: +- Account or partner: +- Segment: +- Owner: +- Start date: +- End date: +- Commercial path: annual license, hosted subscription, validation service, or + strategic evaluation +- Contract value: +- Renewal or expansion signal: + +## Dataset Metadata + +- Dataset source: +- Permission to use: +- Rows: +- Items: +- Response type: +- Missingness summary: +- Covariates or grouping fields: +- Privacy constraints: +- Storage decision: local-only, hosted transient, or explicit export + +## Run Configuration + +- `kaefa` version or commit: +- Runtime surface: R API, Shiny Studio, container, or hosted deployment +- Minimum factors: +- Maximum factors: +- Rotation: +- Model-selection criteria: +- Hardware or runner: + +## Results + +- Report generated: yes, partial, or no +- Time to report: +- Selected factor count: +- Items removed or flagged: +- Fit indices available: +- Exported artifacts: +- Structured failure reason, if any: +- Analyst intervention: none, light, heavy, or impossible + +## Customer Outcome + +- Pilot score: green, yellow, or red +- Buyer-readable value: +- Main blocker: +- Requested product change: +- Willingness to pay: +- Reference permission: public, private, internal only, or none + +## Evidence Links + +- Signed scope or approval: +- Aggregate run note: +- Exported report: +- Pilot scorecard: +- Follow-up decision: + +## Signoff + +- Product owner: +- Statistical reviewer: +- Security/privacy reviewer: +- Commercial owner: +- Date: diff --git a/docs/diligence/release-diligence-checklist.md b/docs/diligence/release-diligence-checklist.md new file mode 100644 index 0000000..63107d4 --- /dev/null +++ b/docs/diligence/release-diligence-checklist.md @@ -0,0 +1,60 @@ +# Release Diligence Checklist + +Use this checklist before presenting `kaefa` as a buyer-facing product, +enterprise pilot, or 2B KRW acquisition candidate. + +## Product Scope + +- Data room index points to current evidence and unresolved decisions. +- Current public APIs are listed and intentionally preserved. +- `kaefa-core`, `kaefa-studio`, and `kaefa-runner` boundaries are documented. +- Known out-of-scope features are named. +- User-facing workflows have success and failure states. + +## Install And Deployment Proof + +- R package installation works from the target branch. +- Shiny launch path works with documented dependencies. +- Hosted or container deployment path is tested when offered to buyers. +- Runtime dependency versions are recorded. + +## Benchmark Evidence + +- `inst/benchmarks/manifest.csv` passes schema validation. +- Each benchmark dataset has source and license context. +- Expected factor-count ranges are reviewed. +- Runtime measurements name hardware or CI runner class. +- Heavy benchmarks are separated from fast PR tests. +- Buyer-facing release evidence is recorded in + `docs/diligence/release-evidence-log.md`. + +## Security And Privacy + +- No secrets, SSH keys, or customer data are committed. +- Uploaded Shiny data is not persisted unless the user exports it. +- Remote execution setup documents SSH key handling. +- Dependency review and security scans are tracked. + +## License And IP + +- `DESCRIPTION` and `LICENSE` agree. +- Dependency license implications are documented. +- Copyright ownership and contributor rights are reviewed. +- Commercial posture is explicit: GPL-only, dual-license, hosted service, or + open-core. + +## Revenue And Pilots + +- Pricing path is selected. +- Paid pilots are scored with a consistent scorecard. +- Each pilot has a filled evidence note based on + `docs/diligence/pilot-evidence-template.md`. +- ARR needed for 2B KRW valuation is stated. +- Renewal, expansion, or strategic buyer path is documented. + +## Known Limitations + +- Missing dependencies are listed. +- Unsupported data shapes are listed. +- Non-convergence behavior is documented. +- Legal or owner decisions are marked separately from engineering blockers. diff --git a/docs/diligence/release-evidence-log.md b/docs/diligence/release-evidence-log.md new file mode 100644 index 0000000..f3f3270 --- /dev/null +++ b/docs/diligence/release-evidence-log.md @@ -0,0 +1,51 @@ +# Release Evidence Log + +Use this log before any buyer-facing release, pilot handoff, or 20억 KRW sale +readiness claim. Each row should point to reproducible evidence rather than a +chat summary. + +## Release Candidate + +- Branch: +- Commit: +- PR: +- Release date: +- Evidence owner: + +## Required Evidence + + + +| Area | Required evidence | Current evidence link | Status | +| --- | --- | --- | --- | +| Package install | `R CMD INSTALL .` on target branch | Runtime-code evidence: Dockerfile step 8 at `b5bfdb8b509eb8ec06f143c6435f880eda3d2e20` completed `* DONE (kaefa)` during `podman build -t kaefa-studio:local .` on 2026-07-03 KST. | captured | +| Fast PR gate | `test-fast` check on current head | | missing | +| R CMD check | Required R-CMD-check matrix on current head | | missing | +| Benchmark manifest | `test-benchmark-manifest.R` result | | missing | +| Shiny surface | `test-shiny-product-surface.R` result | | missing | +| Container | Container build and launch smoke test | Runtime-code evidence: `podman build -t kaefa-studio:local .` produced image `04840d3aa188`; local run on `127.0.0.1:3838` returned HTTP 200, 12,017 bytes, title `kaefa: Automated Exploratory Factor Analysis`, and log `Listening on http://0.0.0.0:3838`. Rerun on the target release commit before buyer handoff. | captured | +| Report export | Human-readable report generated | | missing | +| Runtime | Time-to-report measurement for benchmark class | | missing | +| License/IP | Owner/legal posture reviewed | | missing | +| Privacy | Upload persistence and export behavior reviewed | | missing | +| Security | Dependency/security checks reviewed | | missing | +| Commercial | Pilot or ARR evidence linked | | missing | + + + +## Blocker Rules + +- Review approval is not a product blocker. +- Queued GitHub checks are monitoring state, not a blocker. +- Failed required checks are blockers until logs are reviewed and fixed. +- Missing owner/legal authority is a sale-claim blocker. +- Missing paid-pilot or ARR evidence blocks a revenue-multiple valuation claim. +- Missing runtime evidence blocks hosted-operation claims. + +## Signoff + +- Engineering: +- Product: +- Security/privacy: +- Legal/IP: +- Commercial: diff --git a/docs/diligence/repository-boundary-decision.md b/docs/diligence/repository-boundary-decision.md new file mode 100644 index 0000000..90c8ccf --- /dev/null +++ b/docs/diligence/repository-boundary-decision.md @@ -0,0 +1,115 @@ +# Repository Boundary Decision + +Decision date: 2026-07-03 + +## Decision + +Keep `kaefa` in one repository for the current sale-readiness phase. + +Use explicit internal boundaries instead of a repository split: + +- `kaefa-core`: R package APIs, statistical engine, model selection, + benchmark evidence, and package-level tests. +- `kaefa-studio`: bundled Shiny UI, report export, user-facing validation, + and pilot workflow evidence. +- `kaefa-runner`: future container, hosted execution, remote job control, + monitoring, and support evidence. + +Do not introduce a git submodule now. If a split becomes necessary, prefer a +normal repository or package split with its own release process. Use a submodule +only when a buyer explicitly requires vendored source integration. + +## Rationale + +The current buyer evidence is most useful when product behavior, package tests, +release checks, license posture, deployment examples, and pilot evidence remain +reviewable in one pull request and one data room index. + +A submodule would add operational friction without improving the immediate +diligence story. The current commercial risk is not source layout. The risk is +proving repeatable analysis, pilot value, deployment posture, and IP/license +authority. + +## Options Considered + +### Monorepo With Internal Boundaries + +This is the selected option for the current phase. + +It keeps the R package, Shiny product surface, deployment example, CI gates, and +diligence documents together. That is appropriate while the product is still +being packaged as a single pilot-ready offering. + +### Separate `kaefa-core` Package Or Repository + +Use this after the statistical engine has a stable public API, an independent +release cadence, and at least two consumers beyond the bundled Studio UI. + +This is also appropriate if a buyer wants to license the engine separately from +the UI and deployment surface. + +### Separate `kaefa-studio` Application Repository + +Use this after the Studio surface needs authentication, tenant isolation, +non-R frontend build tooling, independent deployment, or its own release notes. + +Until then, a separate app repository would mostly duplicate packaging and +review work. + +### Separate `kaefa-runner` Service Repository + +Use this after remote execution has queue semantics, resource isolation, audit +logging, service-level monitoring, and support commitments. + +The current container and ShinyProxy example are deployment evidence, not a +separate service boundary yet. + +### Git Submodule + +Avoid this by default. + +A submodule is only justified when a downstream buyer or partner requires +vendored source integration into an existing controlled repository. It should +not be used as the first split mechanism for normal productization. + +## Split Triggers + +Revisit the decision when one or more of these becomes true: + +- a non-R frontend build pipeline is introduced, +- authentication or tenant isolation becomes required, +- Studio and core package releases need different schedules, +- runtime infrastructure needs a separate service-level agreement, +- license or commercial terms differ between engine, UI, and runner, +- at least two external consumers use `kaefa-core` without `kaefa-studio`, +- a buyer requires a separate deployable source package, +- a buyer requires vendored source integration. + +## Current File Boundaries + +- Core package and statistical engine: + `R/`, `NAMESPACE`, `man/`, `tests/testthat/`, + `inst/benchmarks/manifest.csv`, and `docs/validation/`. +- Studio product surface: + `inst/shiny-app/`, `docs/product/kaefa-studio-requirements.md`, + report export behavior, and pilot workflow documentation. +- Runner and deployment evidence: + `Dockerfile`, `.dockerignore`, `deploy/`, + and `docs/operations/deployment.md`. +- Sale-readiness and buyer evidence: + `docs/business/`, `docs/diligence/`, and the FigJam roadmap. + +## Migration Path + +1. Keep the monorepo while closing the current buyer evidence gaps. +2. Make `kaefa-core` boundaries explicit in tests, benchmarks, and examples. +3. Maintain the package-level public API contract in + `docs/product/kaefa-core-api-contract.md` before moving engine code. +4. Split `kaefa-studio` only after it needs independent deployment or auth. +5. Split `kaefa-runner` only after it operates as a service boundary. +6. Use a submodule only for a buyer-required vendored integration. + +## Review Cadence + +Review this decision before any major hosted pilot, buyer technical diligence +handoff, or repository-level license change. diff --git a/docs/operations/deployment.md b/docs/operations/deployment.md new file mode 100644 index 0000000..fb1f60f --- /dev/null +++ b/docs/operations/deployment.md @@ -0,0 +1,70 @@ +# Deployment + +## Local Container Smoke Test + +These commands define the evaluation smoke path. Runtime-code evidence for PR +`#61` was captured on 2026-07-03 KST at commit +`b5bfdb8b509eb8ec06f143c6435f880eda3d2e20`. Rerun this path for release-candidate +signoff and after changes to `Dockerfile`, `DESCRIPTION`, `R/`, or +`inst/shiny-app/`. + +Build the local image: + +```bash +podman build -t kaefa-studio:local . +``` + +Run the Shiny app on port 3838: + +```bash +podman run --rm -p 3838:3838 kaefa-studio:local +``` + +Then open: + +```text +http://localhost:3838 +``` + +Docker can be used instead of Podman with the same `build` and `run` +arguments. + +Runtime-code evidence: + +- `podman build -t kaefa-studio:local .` completed with exit code 0 and + produced image `04840d3aa188`. +- `podman run --rm -d --name kaefa-studio-smoke -p 127.0.0.1:3838:3838 + kaefa-studio:local` launched the app. +- `curl http://127.0.0.1:3838/` returned HTTP 200 with 12,017 bytes and title + `kaefa: Automated Exploratory Factor Analysis`. +- Container logs included `Listening on http://0.0.0.0:3838`. + +## ShinyProxy Example + +`deploy/shinyproxy/application.yml.example` provides a minimal ShinyProxy +configuration for an internal evaluation deployment. +It relies on the Dockerfile `CMD` for the app startup command. + +Before using it outside a local pilot: + +- set `KAEFA_SHINYPROXY_ANALYST_PASSWORD` from the deployment secret manager, +- set memory and CPU limits from measured benchmark runs, +- decide whether uploaded data must stay local-only, +- add HTTPS and organization authentication. + +## Runtime Notes + +The container installs the current source package and runs: + +```r +kaefa::launchAEFA(host = "0.0.0.0", port = 3838) +``` + +The app must not persist uploaded datasets unless a future feature explicitly +adds reviewed storage behavior. + +## Current Limits + +- The image is intended for evaluation, not hardened production. +- Runtime cost and memory requirements still need benchmark measurements. +- Authentication in the ShinyProxy example is a placeholder. diff --git a/docs/product/kaefa-core-api-contract.md b/docs/product/kaefa-core-api-contract.md new file mode 100644 index 0000000..7cfbdb0 --- /dev/null +++ b/docs/product/kaefa-core-api-contract.md @@ -0,0 +1,186 @@ +# kaefa Core API Contract + +Decision date: 2026-07-03 + +## Purpose + +This document defines the current `kaefa-core` API surface for pilot, +buyer-diligence, and future library-split decisions. + +It is based on the exported `NAMESPACE` entries and source-level entry points +reviewed in `R/kaefa.R`, `R/newEngine.R`, and `R/utils.R`. + +## Current Contract Status + +`kaefa-core` is not a separate package yet. It is the internal boundary for the +R statistical engine inside this repository. + +The contract is strong enough for pilot documentation and buyer review, but not +yet strong enough for a separate repository or separately licensed package. + +## Pilot-Facing Public Functions + +These functions are acceptable to describe in pilot and buyer material. + +### `aefa()` + +Primary automated exploratory factor analysis workflow. + +Expected input: + +- numeric response `data.frame` or supported `mirt` model object, +- optional covariate data for fixed and random effects, +- factor extraction bounds, +- model-selection and item-fit controls. + +Expected output: + +- an `aefa` history object when `saveModelHistory = TRUE`, +- otherwise the selected estimated model. + +Contract notes: + +- The search is greedy and heuristic, not a global optimizer. +- Runtime depends on item count, sample count, factor range, model family, and + remote execution settings. +- Buyer-facing reports must preserve selected options, data shape, run time, + and model-selection criteria. + +### `aefaResults()` + +Human-readable model summary surface for an `aefa` result or supported `mirt` +model. + +Contract notes: + +- It may print fit statistics, loadings, and reliability diagnostics. +- It can return a converted model when `returnModel = TRUE`. +- It is a reporting helper, not a serialization format. + +### `evaluateItemFit()` + +Item-fit diagnostic helper for supported `mirt` models and `aefa` results. + +Contract notes: + +- It may compute `Zh`, `S_X2`, `PV_Q1`, infit, and outfit where available. +- Some fit statistics can be unavailable for a given model family. +- Missing diagnostics should be reported as unavailable, not silently treated + as passing evidence. + +### `recursiveFormula()` + +Score or theta extraction helper for fitted models. + +Contract notes: + +- It can return expected test scores or theta estimates. +- `divide` is the preferred argument. The deprecated `devide` spelling is only + a compatibility path. +- Downstream integrations should treat the return shape as dependent on model + type and `individual` or `extractThetaOnly` options. + +## Studio And Runner Entry Points + +These functions belong near `kaefa-core`, but their main contract is product or +runtime packaging rather than statistical estimation. + +### `launchAEFA()` + +Launches the bundled Shiny app from `inst/shiny-app/`. + +Contract notes: + +- This is the `kaefa-studio` entry point. +- It should fail early when the installed package lacks the Shiny app files. +- It should not be used as evidence that hosted authentication, tenant + isolation, or monitoring exists. + +### `aefaInit()` + +Initializes optional local or remote parallel execution. + +Contract notes: + +- This is a `kaefa-runner` precursor, not a production runner contract. +- SSH keys, server names, and remote execution settings must not be committed + to the repository. +- Hosted or managed operation requires a separate runner/service design. + +## Advanced Engine Functions + +These exports are useful for advanced users and tests, but should not be +positioned as the stable buyer-facing API until more contract tests exist. + +### `engineAEFA()` + +Lower-level model candidate estimation engine used by `aefa()`. + +Contract notes: + +- Accepts detailed model, estimation, random-effect, resampling, and + acceleration controls. +- Returns candidate models for selection by the higher-level workflow. +- Should remain available for research users, but buyer-facing workflows should + prefer `aefa()` unless a pilot explicitly needs engine-level control. + +### Theta-Prior Helpers + +Current helpers: + +- `fitThetaPrior()` +- `testThetaPriorCalibration()` +- `applyThetaPrior()` + +Contract notes: + +- These are exploratory calibration helpers. +- They can support product differentiation after validation evidence exists. +- They need dedicated tests and generated `.Rd` documentation before being + marketed as a stable commercial feature. + +## Export Cleanup Candidates + +The current `NAMESPACE` exports several dot-prefixed helpers: + +- `.covdataClassifieder` +- `.covdataFixedEffectComb` +- `.exportParmsEME` +- `.mirt` +- `.mixedmirt` + +These should be treated as internal implementation details for buyer material. + +Before a separate `kaefa-core` package or proprietary API claim, decide whether +to: + +- keep them exported and document them as advanced extension points, +- deprecate them with a compatibility window, +- or stop exporting them in a breaking major release. + +## Minimum Contract Tests Before Split + +A separate core package or repository should not be created until these tests +exist and pass in CI: + +- `aefa()` returns an `aefa` history object for a small known numeric dataset. +- `aefaResults()` handles an `aefa` object without throwing. +- `evaluateItemFit()` reports unavailable diagnostics explicitly. +- `recursiveFormula()` covers both score output and theta-only output. +- `launchAEFA()` fails clearly when Shiny files are unavailable. +- theta-prior helpers cover fit success, unsupported distribution, and fallback + behavior. +- dot-prefixed helpers are either documented or covered by deprecation tests. + +## Split Readiness Rule + +Do not split `kaefa-core` into a separate package or repository until all are +true: + +- stable public API contract is represented by tests, +- exported helper policy is decided, +- release notes distinguish public API changes from internal changes, +- at least two consumers need `kaefa-core` without `kaefa-studio`, +- license and commercial packaging posture is approved by the owner. + +Until then, keep `kaefa-core` as an internal boundary in this repository. diff --git a/docs/product/kaefa-studio-requirements.md b/docs/product/kaefa-studio-requirements.md new file mode 100644 index 0000000..c2b4976 --- /dev/null +++ b/docs/product/kaefa-studio-requirements.md @@ -0,0 +1,80 @@ +# kaefa Studio Requirements + +## Purpose + +`kaefa-studio` is the buyer-facing product surface for researchers, assessment +teams, and consultants who need automated exploratory factor analysis without +writing R code. The current implementation is the bundled Shiny app in +`inst/shiny-app/app.R`; this document defines the minimum product behavior +before any separate app repository or submodule is justified. + +## Primary Workflow + +1. Upload a numeric CSV response dataset. +2. Validate row count, item count, column types, and factor range. +3. Choose minimum factors, maximum factors, rotation, and model-selection + criteria. +4. Run `kaefa::aefa()` with visible progress and clear runtime expectations. +5. Inspect model summary, item fit, factor loadings, and fit indices. +6. Export an RDS result object. +7. Export a human-readable report. +8. Export a reproducibility bundle in a future release. + +## Required Failure States + +### Non-Numeric Columns + +Show a blocking error that names the invalid columns and explains that item +response data must be numeric before factor analysis can run. + +### Factor Count Exceeds Item Count + +Block analysis before calling `aefa()` and tell the user the maximum allowed +factor count for the uploaded dataset. + +### Missing Runtime Dependencies + +If `shiny`, `DT`, or package dependencies are unavailable, the app launch path +must fail before user upload with an install-focused message. + +### Long-Running Model + +After analysis starts, keep an obvious running state visible. If a timeout is +introduced later, return a structured timeout result instead of discarding +partial diagnostic context. + +### Model Non-Convergence + +Return a structured failure reason, the last attempted model context when safe, +and suggested next actions such as reducing maximum factors, checking low +variance items, or switching to a smaller pilot dataset. + +## Buyer-Facing Requirements + +- The first screen must make the product task clear: upload assessment data, + configure automated factor search, and export a report. +- Reports must explain what was selected, why it was selected, and what items + were removed or flagged. +- The app must avoid storing uploaded data unless the user explicitly exports a + result. +- Every exported report should include package version, run timestamp, selected + options, and data shape. +- Default settings should be safe for a first pilot dataset, not optimized for + maximum search breadth. + +## Split Criteria + +Do not split `kaefa-studio` into a separate repository yet. +The repository-level decision is recorded in +`docs/diligence/repository-boundary-decision.md`. + +Split it only when at least one of these is true: + +- it needs authentication or tenant isolation, +- it needs hosted deployment independent of the R package release, +- it adds non-R frontend build tooling, +- it needs independent issue tracking and release notes, +- a buyer specifically requires a separate deployable source package. + +Prefer a normal repository or package split over a git submodule. Use a +submodule only when a buyer explicitly requires vendored source integration. diff --git a/docs/superpowers/plans/2026-07-02-kaefa-2b-krw-sale-readiness.md b/docs/superpowers/plans/2026-07-02-kaefa-2b-krw-sale-readiness.md new file mode 100644 index 0000000..39f15fb --- /dev/null +++ b/docs/superpowers/plans/2026-07-02-kaefa-2b-krw-sale-readiness.md @@ -0,0 +1,587 @@ +# kaefa 2B KRW Sale Readiness Plan + +> **For agentic workers:** REQUIRED SUB-SKILL: Use +> superpowers:subagent-driven-development (recommended) or +> superpowers:executing-plans to implement this plan task-by-task. Steps use +> checkbox (`- [ ]`) syntax for tracking. + +**Goal:** Turn `kaefa` from a research-grade R package plus bundled Shiny app +into a diligence-ready product candidate that can support a 2B KRW acquisition +case. In Korean valuation terms this is the 20억 KRW target from the request, +not a 20B KRW target. + +**Architecture:** Keep one repository for the next execution phase, but enforce +clear package boundaries: `kaefa-core` as the R/statistical engine boundary, +`kaefa-studio` as the product UI boundary, and `kaefa-runner` as the future +execution/deployment boundary. Do not introduce a submodule until the UI or +runner has an independent release cadence, dependency stack, and buyer-facing +deployment target. + +**Tech Stack:** R package, `mirt`, `future`, `testthat`, Shiny, DT, GitHub +Actions, FigJam/Figma diagrams without Figma Code Connect. + +## Global Constraints + +- Preserve current public R APIs unless a task explicitly changes an interface: + `aefa()`, `engineAEFA()`, `aefaInit()`, `aefaResults()`, + `launchAEFA()`, `fitThetaPrior()`, `testThetaPriorCalibration()`, and + `applyThetaPrior()`. +- Use `README.Rmd` as the source for `README.md`; regenerate both together. +- Keep secrets, SSH keys, customer data, and assessment data out of git. +- Treat review approval as non-blocking for this roadmap. Treat broken CI, + failing reproducible tests, missing legal/license authority, or unavailable + runtime dependencies as real blockers. +- Treat queued GitHub checks as monitoring state, not a blocker. Do not bypass + a failed required check. +- Do not use Figma Code Connect for this work. + +--- + +## Evidence Snapshot + +- Repository: `ContextualWisdomLab/kaefa`, public, default branch `develop`. +- Initial baseline head audited: + `334a5484c5e1d0c35e3f8e575f0dc6eb29c39da0`. Current release-candidate + evidence is tracked in `docs/diligence/release-evidence-log.md` and PR `#61`. +- Live open PRs: Dependabot updates `#55`, `#57`, `#58`, `#59`, `#60`; all + blocked by review requirement, not by a proven technical failure. +- Live open issues relevant to productization: GPU integration `#49`, model + comparison via `nonnest2` `#48`, parameter calibration with benchmark datasets + `#46`, and full lint/refactor `#45`. +- Product surface today: R package API, bundled Shiny app in + `inst/shiny-app/app.R`, documentation in `PRD.md`, `TRD.md`, + `ARCHITECTURE.md`, README/vignettes, and a broad test suite. +- Initial local validation snapshot: R 4.6.0 was available; `mirt` and + `testthat` were installed; host-R Shiny dependencies needed explicit + installation outside the container/runtime path. +- CI gap today: `.github/workflows/R-CMD-check.yaml` runs + `check-r-package` with `args: 'c("--no-manual", "--no-tests")'`, so the large + test suite is not currently a required package-check gate. +- License risk today: `DESCRIPTION` declares `GPL-3 + file LICENSE`, but + `LICENSE` currently contains only `NA`. Dependency license review also matters + because `mirt` is GPL (>= 3). +- Figma output: FigJam roadmap created at + . +- Continued Figma output: secondary scratch FigJam board created at + . +- ContextualWisdomLab organization resources reviewed: `.github`, `aFIPC`, + `fast-mlsirm`, `nonnest2`, `noema`, `appguardrail`, and other active product + repositories. See `docs/diligence/contextualwisdomlab-resource-map.md`. + +## Product Decision + +The correct first move is not a submodule split. The current repository is still +small enough that a submodule would add coordination overhead without solving +the buyer-facing risks: validation evidence, deployability, supportability, +license clarity, and revenue proof. + +Use this staged boundary instead: + +1. **Now:** keep a monorepo and document internal boundaries. +2. **After validation gates pass:** move reusable statistical orchestration into + a narrower `kaefa-core` R-package boundary inside `R/`. +3. **After a hosted or enterprise UI exists:** split `kaefa-studio` into a + separate app repo or package if it needs independent deployment, auth, + monitoring, or non-R frontend dependencies. +4. **After remote execution has real operational semantics:** create + `kaefa-runner` as a separate service or deployment package, not a submodule, + unless a buyer specifically requires source-vendored integration. + +## 2B KRW Valuation Frame + +This plan treats 2B KRW as an acquisition value target, not as a claim that the +current repository is worth 2B KRW today. + +The practical target is one of: + +- **Revenue-multiple path:** 400M-700M KRW ARR with credible retention and gross + margin, assuming roughly 3x-5x ARR valuation multiples for a niche private + software asset. +- **Strategic-IP path:** fewer customers, but strong validation evidence, + defensible benchmark results, and a buyer that needs automated IRT/EFA + capability inside a broader assessment platform. +- **Services-to-product path:** 3-5 paid pilots converted into reusable product + workflows, with implementation services explicitly separated from repeatable + software revenue. + +The revenue-multiple path is the cleanest buyer story. The strategic-IP path is +the fallback if the market is specialized but the algorithmic validation is +strong. + +Public-market multiple sources and repository evidence must stay separate: +`docs/business/2b-krw-commercial-model.md` owns valuation assumptions, while +`docs/diligence/contextualwisdomlab-resource-map.md` owns adjacent organization +asset evidence. + +## KPI Framework + +Primary sale-readiness KPIs: + +- **Activated analysis:** percent of new datasets that produce an exported report + without maintainer intervention. +- **Time to report:** median wall-clock time from CSV upload or R API call to a + buyer-readable result package. +- **Benchmark agreement:** percent of benchmark datasets where selected factor + count/model choice matches a reviewed expected result or documented acceptable + range. +- **Model failure explainability:** percent of failed analyses that return a + structured reason and next action instead of silent `NULL` or raw R errors. +- **Paid pilot conversion:** paid pilots signed, completed, and converted to + recurring license or annual support. + +Guardrails: + +- No customer data leaves the user-controlled runtime without explicit + configuration. +- No uploaded Shiny dataset is persisted beyond the session unless the user + explicitly exports it. +- Dependency and license posture must be explainable to a buyer. +- CI must prove the fast test suite on every PR and the heavier benchmark suite + on schedule or release. + +## Task 1: License And Diligence Baseline + +**Files:** + +- Modify: `LICENSE` +- Modify: `DESCRIPTION` +- Create: `docs/diligence/license-and-ip.md` + +**Interfaces:** + +- Consumes: current `DESCRIPTION` license field and dependency list. +- Produces: a buyer-readable license/IP note and a corrected repository license + posture. + +- [x] **Step 1: Decide sale posture** + + Record whether the sellable asset remains GPL-only, becomes dual-licensed, or + separates open-source core from commercial UI/support. If no legal authority + is available, mark this task blocked on owner/legal input instead of guessing. + +- [x] **Step 2: Fix the license file** + + If GPL-3 remains the current posture, replace `LICENSE` with the full GPL-3 + text or the correct CRAN-compatible license file expected by R packaging. + +- [x] **Step 3: Document dependency license implications** + + Create `docs/diligence/license-and-ip.md` with: + + ```md + # License And IP Diligence + + ## Current package license + + kaefa declares `GPL-3 + file LICENSE` in `DESCRIPTION`. + + ## Dependency constraints + + - `mirt`: GPL (>= 3) + - `future`: LGPL (>= 2.1) + - `shiny`: required for UI runtime + - `DT`: required for UI tables + - `fitdistrplus`: required for theta-prior utilities + + ## Sale-readiness decision + + The current working posture is GPL core plus commercial services and hosted + deployment, until owner/legal review explicitly approves a dual-license or + open-core sale structure. + + ## Required owner/legal checks + + - Confirm copyright ownership for all substantial contributions. + - Confirm whether GPL dependencies constrain proprietary distribution. + - Confirm buyer deliverables: source sale, hosted service, or support contract. + ``` + +- [x] **Step 4: Verify** + + Run: + + ```bash + Rscript -e 'read.dcf("DESCRIPTION")[1,"License"]' + ``` + + Expected: prints the selected license field without parsing errors. + +## Task 2: Fast, Required CI Test Gate + +**Files:** + +- Preserve: `.github/workflows/R-CMD-check.yaml` +- Create: `.github/workflows/test-fast.yaml` +- Create: `tests/FAST_TESTS.md` + +**Interfaces:** + +- Consumes: existing `tests/testthat/*.R` suite. +- Produces: a PR-required fast test gate that does not depend on full benchmark + runtime. + +- [x] **Step 1: Keep R CMD check but stop pretending it runs tests** + + Leave `R-CMD-check` as package-install validation if runtime is too heavy, but + add a separate workflow named `test-fast` that runs explicit targeted tests. + +- [x] **Step 2: Add fast workflow** + + Create `.github/workflows/test-fast.yaml`: + + + + ```yaml + name: test-fast + + on: + pull_request: + branches: [main, master, develop] + push: + branches: [main, master, develop] + + permissions: + contents: read + + jobs: + test-fast: + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2 + - uses: r-lib/actions/setup-r@6f6e5bc62fba3a704f74e7ad7ef7676c5c6a2590 # v2 + with: + use-public-rspm: true + - uses: r-lib/actions/setup-r-dependencies@6f6e5bc62fba3a704f74e7ad7ef7676c5c6a2590 # v2 + with: + extra-packages: any::testthat + needs: check + - name: Run fast productization tests + run: | + Rscript - <<'RSCRIPT' + reporter <- testthat::StopReporter$new() + testthat::test_file("tests/testthat/test-benchmark-manifest.R", + reporter = reporter) + testthat::test_file("tests/testthat/test-shiny-product-surface.R", + reporter = reporter) + RSCRIPT + ``` + + + +- [x] **Step 3: Document heavy suite policy** + + Create `tests/FAST_TESTS.md` to state which tests are fast PR gates and + which tests are scheduled/release benchmark gates. + +- [x] **Step 4: Verify** + + Run: + + ```bash + npx -y markdownlint-cli2@0.20.0 tests/FAST_TESTS.md + ``` + + Expected: no markdownlint errors. + +## Task 3: Benchmark Corpus And Validation Harness + +**Files:** + +- Create: `inst/benchmarks/README.md` +- Create: `inst/benchmarks/manifest.csv` +- Create: `tests/testthat/test-benchmark-manifest.R` +- Create: `docs/validation/benchmark-protocol.md` + +**Interfaces:** + +- Consumes: current `aefa()` and `engineAEFA()` APIs. +- Produces: a repeatable evidence base for buyer diligence. + +- [x] **Step 1: Define benchmark manifest** + + Create `inst/benchmarks/manifest.csv` with columns: + + ```csv + dataset_id,source,license,rows,items,response_type,expected_factor_min,expected_factor_max,expected_runtime_seconds,notes + science,mirt::Science,mirt,392,4,mixed,1,2,120,Smoke + ``` + +- [x] **Step 2: Add manifest test** + + Create `tests/testthat/test-benchmark-manifest.R`: + + ```r + test_that("benchmark manifest has required columns", { + manifest <- read.csv(test_path("../../inst/benchmarks/manifest.csv")) + expect_true(all(c( + "dataset_id", "source", "license", "rows", "items", "response_type", + "expected_factor_min", "expected_factor_max", + "expected_runtime_seconds", "notes" + ) %in% names(manifest))) + expect_false(anyDuplicated(manifest$dataset_id)) + }) + ``` + +- [x] **Step 3: Write protocol** + + `docs/validation/benchmark-protocol.md` must define: + + - dataset admission criteria, + - expected-result review process, + - runtime measurement method, + - acceptable failure categories, + - release-signoff requirements. + +- [x] **Step 4: Verify** + + Run: + + ```bash + Rscript -e 'testthat::test_file("tests/testthat/test-benchmark-manifest.R")' + ``` + + Expected: one passing test file. + +## Task 4: Product UI Boundary + +**Files:** + +- Modify: `inst/shiny-app/app.R` +- Create: `docs/product/kaefa-studio-requirements.md` + +**Interfaces:** + +- Consumes: existing Shiny app and `launchAEFA()`. +- Produces: a clear `kaefa-studio` product surface without splitting repos yet. + +- [x] **Step 1: Document buyer-facing UI workflow** + + Create `docs/product/kaefa-studio-requirements.md` with the minimum workflow: + upload numeric CSV, validate items, choose factor range, run analysis, inspect + progress, export report, export reproducibility bundle. + +- [x] **Step 2: Add failure-state requirements** + + Specify copy and behavior for: + + - non-numeric columns, + - factor count greater than item count, + - missing `shiny` or `DT`, + - long-running model timeout, + - model convergence failure. + +- [x] **Step 3: Defer repo split** + + Add a section named `Split Criteria`: + + - split when UI needs auth, tenancy, hosted deployment, or non-R frontend + build tooling; + - do not use a submodule unless a downstream buyer requires vendored source. + +- [x] **Step 4: Verify** + + Run: + + ```bash + npx -y markdownlint-cli2@0.20.0 docs/product/kaefa-studio-requirements.md + ``` + + Expected: no markdownlint errors. + +## Task 5: Runner And Deployment Package + +**Files:** + +- Create: `Dockerfile` +- Create: `deploy/shinyproxy/application.yml.example` +- Create: `docs/operations/deployment.md` + +**Interfaces:** + +- Consumes: `launchAEFA()` and `inst/shiny-app/app.R`. +- Produces: an enterprise-evaluable runtime path. + +- [x] **Step 1: Package the app** + + Add a Dockerfile that installs package dependencies and runs the Shiny app. + +- [x] **Step 2: Add ShinyProxy example** + + Add a ShinyProxy `application.yml.example` that documents image name, app + port, authentication placeholder, resource limits, and no bundled secrets. + +- [x] **Step 3: Document local smoke test** + + `docs/operations/deployment.md` must include: + + ```bash + podman build -t kaefa-studio:local . + podman run --rm -p 3838:3838 kaefa-studio:local + ``` + + Docker can use the same arguments where Docker is the available runtime. + +- [x] **Step 4: Verify** + + Run: + + ```bash + podman build -t kaefa-studio:local . + ``` + + Expected: image builds without missing R package dependencies. + +## Task 6: Revenue And Pilot Evidence + +**Files:** + +- Create: `docs/business/2b-krw-commercial-model.md` +- Create: `docs/business/pilot-scorecard.md` + +**Interfaces:** + +- Consumes: KPI framework in this plan. +- Produces: buyer-readable revenue proof targets and pilot acceptance criteria. + +- [x] **Step 1: Write commercial model** + + Create `docs/business/2b-krw-commercial-model.md` with three pricing paths: + + - annual institution license, + - hosted assessment analytics workspace, + - validation/reporting services attached to product subscription. + +- [x] **Step 2: Set 2B KRW thresholds** + + Include this table: + + ```md + | Exit multiple | ARR needed for 2B KRW value | + | --- | ---: | + | 3x ARR | 667M KRW | + | 4x ARR | 500M KRW | + | 5x ARR | 400M KRW | + | Strategic IP sale | Lower ARR possible with stronger benchmark evidence | + ``` + +- [x] **Step 3: Write pilot scorecard** + + `docs/business/pilot-scorecard.md` must score each pilot on: + + - dataset complexity, + - successful report generation, + - time-to-report, + - analyst intervention required, + - willingness to pay, + - renewal or expansion path, + - security/privacy constraints. + +- [x] **Step 4: Verify** + + Run: + + ```bash + npx -y markdownlint-cli2@0.20.0 docs/business/2b-krw-commercial-model.md docs/business/pilot-scorecard.md + ``` + + Expected: no markdownlint errors. + +## Task 7: Release Diligence Bundle + +**Files:** + +- Create: `docs/diligence/release-diligence-checklist.md` +- Modify: `ARCHITECTURE.md` +- Modify: `TRD.md` + +**Interfaces:** + +- Consumes: all prior tasks. +- Produces: a concise buyer due-diligence package. + +- [x] **Step 1: Add checklist** + + Create `docs/diligence/release-diligence-checklist.md` with sections: + + - product scope, + - install and deployment proof, + - benchmark evidence, + - security and privacy posture, + - license and IP posture, + - revenue and pilots, + - known limitations. + +- [x] **Step 2: Update architecture** + + Update `ARCHITECTURE.md` with the internal boundaries: + `kaefa-core`, `kaefa-studio`, and `kaefa-runner`. + +- [x] **Step 3: Update TRD** + + Update `TRD.md` with the CI/test-gate distinction: + fast PR gate, scheduled benchmark gate, release diligence gate. + +- [x] **Step 4: Verify** + + Run: + + ```bash + npx -y markdownlint-cli2@0.20.0 ARCHITECTURE.md TRD.md docs/diligence/release-diligence-checklist.md + ``` + + Expected: no markdownlint errors. + +## Execution Order + +1. Complete Task 1 first. License/IP ambiguity can invalidate the commercial + story. +2. Complete Task 2 second. A buyer will discount a statistical package whose + tests are present but not enforced. +3. Complete Task 3 third. Benchmark evidence is the core technical moat. +4. Complete Task 4 fourth. The UI needs product-grade failure states before a + non-R buyer can evaluate it. +5. Complete Task 5 fifth. Hosted deployment is optional for research use, but + mandatory for enterprise evaluation. +6. Complete Task 6 sixth. The valuation target needs revenue proof or a clear + strategic-IP rationale. +7. Complete Task 7 last. The diligence bundle is a packaging pass over evidence, + not a substitute for evidence. + +## Current Blockers + +- Runtime-code container build and local Shiny smoke evidence was captured on + 2026-07-03 KST for commit `b5bfdb8b509eb8ec06f143c6435f880eda3d2e20`. + `podman build -t kaefa-studio:local .` completed with exit code 0, produced + image `04840d3aa188`, and `curl http://127.0.0.1:3838/` returned HTTP 200, + 12,017 bytes, and title `kaefa: Automated Exploratory Factor Analysis`. + Rerun for release-candidate signoff and after runtime, Dockerfile, or package + dependency changes. +- Direct host-R Shiny runs still depend on the host installing `shiny`, `DT`, + and `fitdistrplus`. +- CodeGraph tools were not available for this execution environment's R source + analysis, so R implementation checks used native file reads. +- Legal/license posture still needs owner/legal confirmation before changing + away from GPL or claiming proprietary sale rights. + +## References + +- FigJam roadmap: + +- Current market context checked on 2026-07-02: + - [Spherical Insights psychometric tests market] reported the global market + at USD 9.47B in 2023 and forecast USD 30.12B by 2033. + - [Market Research Future online testing software market] reported the market + at USD 5.279B in 2024 and forecast USD 14.87B by 2035. + - [Coherent Market Insights assessment services market] estimated assessment + services at USD 13.238B in 2026 and forecast USD 32.721B in 2033. + - [Aventis Advisors SaaS valuation multiples] reported a March 2026 median + SaaS EV/revenue multiple of 3.4x. + - [SaaS Capital valuation multiples] reported 2025 private SaaS valuation + estimates around 4.8x-5.3x current run-rate annualized revenue. + - `mirt` remains the critical statistical dependency and is GPL (>= 3). + +[Spherical Insights psychometric tests market]: https://www.sphericalinsights.com/reports/psychometric-tests-market +[Market Research Future online testing software market]: https://www.marketresearchfuture.com/reports/online-testing-software-market-39186 +[Coherent Market Insights assessment services market]: https://www.coherentmarketinsights.com/market-insight/assessment-services-market-5935 +[Aventis Advisors SaaS valuation multiples]: https://aventis-advisors.com/saas-valuation-multiples/ +[SaaS Capital valuation multiples]: https://www.saas-capital.com/blog-posts/saas-valuation-multiples-understanding-the-new-normal/ diff --git a/docs/validation/benchmark-protocol.md b/docs/validation/benchmark-protocol.md new file mode 100644 index 0000000..a200a3d --- /dev/null +++ b/docs/validation/benchmark-protocol.md @@ -0,0 +1,76 @@ +# Benchmark Protocol + +## Purpose + +The benchmark corpus turns `kaefa` validation from anecdotal examples into a +repeatable buyer-diligence artifact. Each benchmark should answer one question: +does `kaefa` produce an expected, explainable result within an acceptable +runtime budget for a known class of assessment data? + +## Dataset Admission Criteria + +A dataset can enter the benchmark corpus when all of these are true: + +- its source and license are documented, +- it can be used in CI or a controlled release-validation environment, +- its row count, item count, and response type are recorded, +- an expected factor-count range is reviewed, +- its runtime budget is realistic for the selected validation tier. + +Do not commit customer, private, regulated, or otherwise restricted assessment +data to the repository. + +## Validation Tiers + +- **Manifest tier:** validates benchmark metadata shape and basic ranges on + every PR. +- **Smoke tier:** runs fast installed-package or example-data checks on PRs when + dependencies are available. +- **Release tier:** runs heavier model-fitting checks before tagged releases or + buyer-facing diligence packages. +- **Private tier:** runs restricted datasets outside public CI, with only + reviewed aggregate evidence committed back to documentation. + +## Expected Result Review + +For each dataset, record the expected factor-count range instead of a single +point estimate unless a reviewed source makes the exact value defensible. A +reviewed range is acceptable when it comes from one of: + +- published dataset documentation, +- a reviewed statistical analysis note, +- repeated historical `kaefa` runs with stable output, +- expert review by the package owner or a psychometrics reviewer. + +## Runtime Measurement + +Record runtime as wall-clock seconds from the start of the `aefa()` call to the +available result object. Runtime budgets should name the hardware or CI runner +class used for measurement when the benchmark graduates from manifest tier. + +## Acceptable Failure Categories + +Failures should be classified before release signoff: + +- missing dependency, +- invalid dataset metadata, +- model non-convergence, +- runtime budget exceeded, +- unsupported response type, +- expected-result mismatch, +- unstructured error or silent `NULL`. + +Unstructured errors and silent `NULL` results are product defects for +sale-readiness purposes, even when the underlying statistical model cannot be +fit. + +## Release Signoff + +A release or buyer-facing diligence package should include: + +- manifest validation results, +- smoke benchmark results, +- release-tier benchmark results when available, +- known benchmark exclusions, +- hardware/runtime notes, +- reviewer and date for any expected-result changes. diff --git a/inst/benchmarks/README.md b/inst/benchmarks/README.md new file mode 100644 index 0000000..db333a5 --- /dev/null +++ b/inst/benchmarks/README.md @@ -0,0 +1,18 @@ +# kaefa Benchmark Corpus + +This directory records datasets used to validate `kaefa` behavior across +releases. The first corpus version is intentionally small: it defines the +manifest contract before adding heavier runtime benchmarks. + +Each benchmark dataset must have an entry in `manifest.csv` with: + +- source and license context, +- dataset shape, +- response type, +- expected factor-count range, +- expected runtime budget, +- notes about why the dataset belongs in the corpus. + +Do not commit private, restricted, or customer-owned assessment data here. +Private validation datasets should live in a separately controlled storage +location and be referenced only by reviewed metadata. diff --git a/inst/benchmarks/manifest.csv b/inst/benchmarks/manifest.csv new file mode 100644 index 0000000..d8e7629 --- /dev/null +++ b/inst/benchmarks/manifest.csv @@ -0,0 +1,2 @@ +dataset_id,source,license,rows,items,response_type,expected_factor_min,expected_factor_max,expected_runtime_seconds,notes +science,mirt::Science,GPL-3-or-later,392,4,mixed,1,2,120,mirt package example data; smoke benchmark for dependency and manifest validation diff --git a/inst/shiny-app/app.R b/inst/shiny-app/app.R index bfc669b..155725a 100644 --- a/inst/shiny-app/app.R +++ b/inst/shiny-app/app.R @@ -177,9 +177,19 @@ server <- function(input, output, session) { values <- reactiveValues( data = NULL, results = NULL, + runOptions = NULL, analysisComplete = FALSE ) + clearAnalysisState <- function(clear_data = FALSE) { + if (clear_data) { + values$data <- NULL + } + values$results <- NULL + values$runOptions <- NULL + values$analysisComplete <- FALSE + } + select_model_index <- function(results) { if (!inherits(results, "aefa")) { return(NULL) @@ -213,9 +223,7 @@ server <- function(input, output, session) { type = "error", duration = 10 ) - values$data <- NULL - values$analysisComplete <- FALSE - values$results <- NULL + clearAnalysisState(clear_data = TRUE) return() } @@ -226,15 +234,14 @@ server <- function(input, output, session) { values$data <- read.csv(input$dataFile$datapath, header = input$hasHeader, stringsAsFactors = FALSE) - if (!all(sapply(values$data, is.numeric))) { + invalid_columns <- kaefa:::.kaefaStudioInvalidColumns(values$data) + if (length(invalid_columns) > 0) { showNotification( - "All columns must be numeric for factor analysis. Please check your CSV file.", + kaefa:::.kaefaStudioInvalidColumnMessage(invalid_columns), type = "error", duration = 10 ) - values$data <- NULL - values$analysisComplete <- FALSE - values$results <- NULL + clearAnalysisState(clear_data = TRUE) return() } } else if (ext == "rds") { @@ -243,21 +250,16 @@ server <- function(input, output, session) { type = "error", duration = 10 ) - values$data <- NULL - values$analysisComplete <- FALSE - values$results <- NULL + clearAnalysisState(clear_data = TRUE) return() } else { showNotification(paste("Unsupported file type:", ext), type = "error", duration = 10) - values$data <- NULL - values$analysisComplete <- FALSE - values$results <- NULL + clearAnalysisState(clear_data = TRUE) return() } - values$analysisComplete <- FALSE - values$results <- NULL + clearAnalysisState(clear_data = FALSE) showNotification("Data loaded successfully!", type = "message") @@ -331,6 +333,8 @@ server <- function(input, output, session) { duration = NULL, id = "analysisProgress", type = "message") tryCatch({ + values$runOptions <- kaefa:::.kaefaStudioRunOptions(values$data, input) + # Run aefa analysis values$results <- kaefa::aefa( data = values$data, @@ -356,7 +360,7 @@ server <- function(input, output, session) { removeNotification(id = "analysisProgress") showNotification(paste("Error during analysis:", e$message), type = "error", duration = 15) - values$analysisComplete <- FALSE + clearAnalysisState(clear_data = FALSE) }) }) @@ -482,9 +486,22 @@ server <- function(input, output, session) { sink(file) on.exit(sink(), add = TRUE) + run_options <- values$runOptions + cat("kaefa: Automated Exploratory Factor Analysis\n") cat("=============================================\n\n") cat("Report generated:", format(Sys.time(), "%Y-%m-%d %H:%M:%S"), "\n\n") + cat( + paste( + kaefa:::.kaefaStudioReportMetadataLines( + run_options, + values$data, + input + ), + collapse = "\n" + ), + "\n\n" + ) if (inherits(values$results, "aefa")) { cat("Analysis Summary:\n") diff --git a/tests/FAST_TESTS.md b/tests/FAST_TESTS.md new file mode 100644 index 0000000..5629ada --- /dev/null +++ b/tests/FAST_TESTS.md @@ -0,0 +1,24 @@ +# Fast Test Gate + +`test-fast` is the lightweight PR gate for metadata and productization checks +that do not require expensive model fitting. + +It currently runs: + +```r +reporter <- testthat::StopReporter$new() +testthat::test_file("tests/testthat/test-benchmark-manifest.R", + reporter = reporter) +testthat::test_file("tests/testthat/test-shiny-product-surface.R", + reporter = reporter) +testthat::test_file("tests/testthat/test-core-api-contract.R", + reporter = reporter) +``` + +`R-CMD-check` remains the package installation and multi-OS compatibility gate. +The current workflow uses `--no-tests`, so it should not be treated as proof +that the full statistical test suite ran. + +Release and buyer-diligence runs should add heavier benchmark tests from +`inst/benchmarks/manifest.csv` once expected results and runtime budgets are +reviewed. diff --git a/tests/testthat/helper-test-data.R b/tests/testthat/helper-test-data.R index 04f072a..b64cb03 100644 --- a/tests/testthat/helper-test-data.R +++ b/tests/testthat/helper-test-data.R @@ -1,4 +1,33 @@ # Shared helpers for test data generation. +.find_kaefa_root <- function() { + candidates <- unique(c( + getwd(), + testthat::test_path("../.."), + testthat::test_path("..") + )) + + for (candidate in candidates) { + if (file.exists(file.path(candidate, "DESCRIPTION"))) { + return(normalizePath(candidate, mustWork = TRUE)) + } + } + + stop("Cannot locate kaefa package root for test bootstrap.", call. = FALSE) +} + +.ensure_kaefa_namespace <- function() { + if (requireNamespace("kaefa", quietly = TRUE)) { + return(invisible(TRUE)) + } + + if (!requireNamespace("pkgload", quietly = TRUE)) { + testthat::skip("kaefa is not installed and pkgload is unavailable") + } + + pkgload::load_all(.find_kaefa_root(), export_all = FALSE, quiet = TRUE) + invisible(TRUE) +} + create_test_data <- function(n_items = 10, n_obs = 100) { set.seed(123) data <- data.frame(matrix( @@ -40,6 +69,7 @@ create_binary_test_data <- function(n_items = 10, n_obs = 100) { aefa <- function(...) { .skip_expensive_ci_calls("aefa") + .ensure_kaefa_namespace() defaults <- list( NCYCLES = 120, BURNIN = 40, @@ -57,6 +87,7 @@ efa <- aefa engineAEFA <- function(...) { .skip_expensive_ci_calls("engineAEFA") + .ensure_kaefa_namespace() defaults <- list( NCYCLES = 120, BURNIN = 40, diff --git a/tests/testthat/test-benchmark-manifest.R b/tests/testthat/test-benchmark-manifest.R new file mode 100644 index 0000000..ff10195 --- /dev/null +++ b/tests/testthat/test-benchmark-manifest.R @@ -0,0 +1,69 @@ +benchmark_manifest_path <- function() { + installed_path <- system.file( + "benchmarks", + "manifest.csv", + package = "kaefa" + ) + + if (nzchar(installed_path) && file.exists(installed_path)) { + return(installed_path) + } + + repo_paths <- c( + file.path("inst", "benchmarks", "manifest.csv"), + file.path("..", "..", "inst", "benchmarks", "manifest.csv") + ) + repo_paths[file.exists(repo_paths)][1] +} + +test_that("benchmark manifest has required columns", { + manifest_path <- benchmark_manifest_path() + + if (is.na(manifest_path) || !nzchar(manifest_path)) { + testthat::fail("benchmark manifest not found") + return(invisible(NULL)) + } + + manifest <- read.csv(manifest_path, stringsAsFactors = FALSE) + required_columns <- c( + "dataset_id", + "source", + "license", + "rows", + "items", + "response_type", + "expected_factor_min", + "expected_factor_max", + "expected_runtime_seconds", + "notes" + ) + + expect_true(all(required_columns %in% names(manifest))) + + numeric_columns <- c( + "rows", + "items", + "expected_factor_min", + "expected_factor_max", + "expected_runtime_seconds" + ) + parsed_numeric <- lapply(manifest[numeric_columns], function(column) { + suppressWarnings(as.numeric(column)) + }) + + expect_true(all(vapply(parsed_numeric, function(column) { + all(!is.na(column)) + }, logical(1)))) + + manifest[numeric_columns] <- parsed_numeric + + expect_equal(anyDuplicated(manifest$dataset_id), 0L) + expect_true(all(nzchar(manifest$dataset_id))) + expect_true(all(manifest$rows > 0)) + expect_true(all(manifest$items > 0)) + expect_true(all(manifest$expected_factor_min >= 1)) + expect_true(all(manifest$expected_factor_min <= manifest$items)) + expect_true(all(manifest$expected_factor_max >= manifest$expected_factor_min)) + expect_true(all(manifest$expected_factor_max <= manifest$items)) + expect_true(all(manifest$expected_runtime_seconds > 0)) +}) diff --git a/tests/testthat/test-core-api-contract.R b/tests/testthat/test-core-api-contract.R new file mode 100644 index 0000000..3203248 --- /dev/null +++ b/tests/testthat/test-core-api-contract.R @@ -0,0 +1,82 @@ +repo_file <- function(...) { + repo_paths <- c( + file.path(...), + file.path("..", "..", ...) + ) + repo_path <- repo_paths[file.exists(repo_paths)][1] + + if (is.na(repo_path) || !nzchar(repo_path)) { + testthat::skip(paste("repository file not found:", file.path(...))) + } + + repo_path +} + +namespace_exports <- function() { + if (!requireNamespace("kaefa", quietly = TRUE)) { + testthat::fail("kaefa namespace not available") + return(character()) + } + + getNamespaceExports("kaefa") +} + +api_contract_lines <- function() { + readLines( + repo_file("docs", "product", "kaefa-core-api-contract.md"), + warn = FALSE + ) +} + +test_that("core API contract covers pilot-facing exports", { + exports <- namespace_exports() + contract <- api_contract_lines() + + pilot_exports <- c( + "aefa", + "aefaResults", + "evaluateItemFit", + "recursiveFormula", + "launchAEFA", + "aefaInit", + "engineAEFA", + "fitThetaPrior", + "testThetaPriorCalibration", + "applyThetaPrior" + ) + + expect_true(all(pilot_exports %in% exports)) + expect_true(all(vapply( + pilot_exports, + function(export_name) { + any(grepl(paste0("`", export_name, "\\(\\)`"), contract)) + }, + logical(1) + ))) +}) + +test_that("dot-prefixed exports are tracked as cleanup candidates", { + exports <- namespace_exports() + contract <- api_contract_lines() + + dot_exports <- exports[startsWith(exports, ".")] + + expect_setequal( + dot_exports, + c( + ".covdataClassifieder", + ".covdataFixedEffectComb", + ".exportParmsEME", + ".mirt", + ".mixedmirt" + ) + ) + + expect_true(all(vapply( + dot_exports, + function(export_name) { + any(grepl(export_name, contract, fixed = TRUE)) + }, + logical(1) + ))) +}) diff --git a/tests/testthat/test-shiny-product-surface.R b/tests/testthat/test-shiny-product-surface.R new file mode 100644 index 0000000..d0cd182 --- /dev/null +++ b/tests/testthat/test-shiny-product-surface.R @@ -0,0 +1,94 @@ +shiny_app_path <- function() { + installed_path <- system.file( + "shiny-app", + "app.R", + package = "kaefa" + ) + + if (nzchar(installed_path) && file.exists(installed_path)) { + return(installed_path) + } + + app_paths <- c( + file.path("inst", "shiny-app", "app.R"), + file.path("..", "..", "inst", "shiny-app", "app.R") + ) + app_path <- app_paths[file.exists(app_paths)][1] + + if (is.na(app_path) || !nzchar(app_path)) { + testthat::fail("Shiny app source file not found") + return(invisible(NULL)) + } + + app_path +} + +test_that("Shiny upload validation identifies invalid columns", { + app_path <- shiny_app_path() + + app_source <- readLines(app_path, warn = FALSE) + sample_data <- data.frame( + item_1 = c(1, 2, 3), + item_2 = c("low", "mid", "high"), + item_3 = c(3, 2, 1) + ) + + invalid_columns <- kaefa:::.kaefaStudioInvalidColumns(sample_data) + validation_message <- kaefa:::.kaefaStudioInvalidColumnMessage( + invalid_columns + ) + + expect_equal(invalid_columns, "item_2") + expect_match(validation_message, "Non-numeric columns: item_2", fixed = TRUE) + + expect_true(any(grepl(".kaefaStudioInvalidColumns", app_source, + fixed = TRUE))) +}) + +test_that("Shiny report includes reproducibility metadata", { + app_path <- shiny_app_path() + + app_source <- readLines(app_path, warn = FALSE) + sample_data <- data.frame( + item_1 = c(1, 2, 3), + item_2 = c(3, 2, 1) + ) + sample_input <- list( + minFactors = 1, + maxFactors = 2, + rotation = "bifactorQ", + modelSelection = "DIC", + saveHistory = TRUE + ) + run_options <- kaefa:::.kaefaStudioRunOptions( + sample_data, + sample_input, + package_version = "0.0.test", + started_at = "2026-07-03T00:00:00Z" + ) + metadata_lines <- kaefa:::.kaefaStudioReportMetadataLines( + run_options, + sample_data, + sample_input + ) + + expect_true(any(grepl("kaefa package version: 0.0.test", + metadata_lines, fixed = TRUE))) + expect_true(any(grepl("Data shape: 3 rows x 2 items", + metadata_lines, fixed = TRUE))) + expect_true(any(grepl("Selected options", metadata_lines, fixed = TRUE))) + + expect_true(any(grepl("runOptions", app_source, fixed = TRUE))) + expect_true(any(grepl(".kaefaStudioReportMetadataLines", app_source, + fixed = TRUE))) +}) + +test_that("Shiny app centralizes analysis state resets", { + app_path <- shiny_app_path() + app_source <- readLines(app_path, warn = FALSE) + + expect_true(any(grepl("clearAnalysisState <- function", app_source, + fixed = TRUE))) + expect_equal(sum(grepl("values$runOptions <- NULL", app_source, + fixed = TRUE)), 1L) +})