diff --git a/LICENSE b/LICENSE
--- a/LICENSE
+++ b/LICENSE
@@ -1,674 +1,20 @@
- 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.
+Copyright (c) 2026 Michael Chavinda
-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".
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
- 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 above copyright notice and this permission notice shall be included
+in all copies or substantial portions of the Software.
- 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
-.
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
+CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -1,177 +1,415 @@
+
+
# dataframe-persistent
-Persistent database integration for the Haskell DataFrame library.
+Load SQLite into a [`dataframe`](https://hackage.haskell.org/package/dataframe). The schema is read
+for you: at runtime when you're exploring, or at compile time when you want type safety. You don't
+write a `persistent` entity, a `persistLowerCase` block, or any instances.
-## Overview
+> **This README is a runnable [scripths](https://github.com/DataHaskell/scripths) notebook.**
+> Every Haskell block runs top-to-bottom in one shared session against
+> [`./data/chinook.db`](./data/chinook.db). Reproduce every output below with
+> `scripths docs/base_scripts/base_readme.md -o README.md` run from `dataframe-persistent/`.
-This package provides seamless integration between the `dataframe` library and the `persistent` database library, allowing you to:
+## Three ways in
-- Load database entities directly into DataFrames
-- Perform DataFrame operations on database data
-- Save DataFrame results back to the database
-- Work with type-safe database entities
+| Tier | You write | You get |
+|------|-----------|---------|
+| **Runtime** | `readTable db "artists"` | a `DataFrame`, types inferred from the schema |
+| **Typed** | `$(declareTable db "artists")` + `readTableTyped @Schema` | a compile-time schema type; columns checked by `col @"Name"` |
+| **Persistent** | `$(declareEntity db "artists")` | a full `persistent` entity: typed `Filter` DSL, write-back |
-## Installation
+## Tier 0: runtime reads
-Add to your `package.yaml`:
+This is the quickest way in, similar to pandas' `read_sql_table` / `read_sql`. You point at a file
+and a table.
-```yaml
-dependencies:
-- dataframe ^>= 0.4
-- dataframe-persistent ^>= 0.1
-- persistent >= 2.14
-- persistent-sqlite >= 2.13 # or your preferred backend
+```haskell
+-- cabal: build-depends: dataframe, dataframe-persistent, text
+-- cabal: default-extensions: OverloadedStrings, TemplateHaskell, DataKinds
+-- cabal: default-extensions: TypeApplications, TypeOperators, FlexibleContexts
+import qualified DataFrame as D
+import DataFrame.IO.Persistent.Read
+import Data.Function ((&))
```
-Or to your `.cabal` file:
+>
-```cabal
-build-depends:
- dataframe ^>= 0.4,
- dataframe-persistent ^>= 0.1,
- persistent >= 2.14,
- persistent-sqlite >= 2.13
+What tables are in this database?
+
+```haskell
+listTables "./data/chinook.db"
```
-## Quick Start
+>
+> ["albums","artists","customers","employees","genres","invoice_items","invoices","media_types","playlist_track","playlists","tracks"]
+`describeTable` shows a table's columns and their inferred types before you load it (it returns a
+`describeColumns`-style `DataFrame`):
+
```haskell
-{-# LANGUAGE DataKinds #-}
-{-# LANGUAGE DerivingStrategies #-}
-{-# LANGUAGE FlexibleInstances #-}
-{-# LANGUAGE GADTs #-}
-{-# LANGUAGE GeneralizedNewtypeDeriving #-}
-{-# LANGUAGE MultiParamTypeClasses #-}
-{-# LANGUAGE OverloadedStrings #-}
-{-# LANGUAGE QuasiQuotes #-}
-{-# LANGUAGE StandaloneDeriving #-}
-{-# LANGUAGE TemplateHaskell #-}
-{-# LANGUAGE TypeApplications #-}
-{-# LANGUAGE TypeFamilies #-}
-{-# LANGUAGE UndecidableInstances #-}
+D.toMarkdown' <$> describeTable "./data/chinook.db" "artists"
+```
-import Control.Monad.IO.Class (liftIO)
-import Database.Persist
-import Database.Persist.Sqlite
-import Database.Persist.TH
-import qualified DataFrame as DF
-import qualified DataFrame.Functions as F
-import DataFrame.IO.Persistent
-import DataFrame.IO.Persistent.TH
-import qualified Data.Vector as V
+>
+> | Column Name
Text | Type
Text | SQLite Type
Text | Nullable
Bool | Primary Key
Bool |
+> | --------------------|--------------|---------------------|------------------|-------------------- |
+> | ArtistId | Int | INTEGER | False | True |
+> | Name | Maybe Text | NVARCHAR(120) | True | False |
-import DataFrame.Functions ((.<))
+Load the whole table. The column types (and nullability) come from the schema:
--- Define your entities
-share [mkPersist sqlSettings, mkMigrate "migrateAll"] [persistLowerCase|
-TestUser
- name Text
- age Int
- active Bool
- deriving Show Eq
-|]
+```haskell
+D.toMarkdown' . D.take 5 <$> readTable "./data/chinook.db" "artists"
+```
--- Derive DataFrame instances
-$(derivePersistentDataFrame ''TestUser)
+>
+> | ArtistId
Int | Name
Maybe Text |
+> | ----------------|------------------------- |
+> | 1 | Just "AC/DC" |
+> | 2 | Just "Accept" |
+> | 3 | Just "Aerosmith" |
+> | 4 | Just "Alanis Morissette" |
+> | 5 | Just "Alice In Chains" |
--- Example usage
-main :: IO ()
-main = runSqlite "example.db" $ do
- -- Run migrations
- runMigration migrateAll
-
- -- Insert some test data
- _ <- insert $ TestUser "Alice" 25 True
- _ <- insert $ TestUser "Bob" 30 False
- _ <- insert $ TestUser "Charlie" 35 True
-
- -- Load from database
- allUsersDF <- fromPersistent @TestUser []
- liftIO $ putStrLn $ "Loaded " ++ show (nRows allUsersDF) ++ " users"
-
- -- Load with filters
- activeUsersDF <- fromPersistent @TestUser [TestUserActive ==. True]
- liftIO $ putStrLn $ "Active users: " ++ show (nRows activeUsersDF)
-
- -- Process with DataFrame operations
- -- Expressions are automaticaly generated.
- let youngUsers = DF.filterWhere (test_user_age .< 30) allUsersDF
- ages = V.toList $ DF.columnAsVector test_user_age youngUsers
- liftIO $ putStrLn $ "Young user ages: " ++ show ages
-
- -- Custom configuration
- let config = defaultPersistentConfig
- { pcIdColumnName = "user_id"
- , pcIncludeId = True
- }
- customDF <- fromPersistentWith @TestUser config []
- liftIO $ putStrLn $ "Columns with custom config: " ++ show (DF.columnNames customDF)
+Tables larger than memory? Filter them in the database. Raw SQL works as you'd expect:
+
+```haskell
+D.toMarkdown' <$> readSql "./data/chinook.db" "SELECT * FROM artists WHERE Name LIKE 'A%' LIMIT 5"
```
-## Features
+>
+> | ArtistId
Int | Name
Text |
+> | ----------------|------------------ |
+> | 1 | AC/DC |
+> | 2 | Accept |
+> | 3 | Aerosmith |
+> | 4 | Alanis Morissette |
+> | 5 | Alice In Chains |
-- **Type-safe conversions** between Persistent entities and DataFrames
-- **Template Haskell support** for automatic instance generation
-- **Configurable loading** with batch size and column selection
-- **Column name cleaning** - removes table prefixes automatically (e.g., `test_user_name` → `name`)
-- **Automatically generate typed expressions** - creates expressions in snake case prefixed by table name (e.g `test_user_name`).
-- **Type preservation** - maintains proper types for Text, Int, Bool, Day, etc.
-- **Empty DataFrame support** - preserves column structure even with no data
-- **Support for all Persistent backends** (SQLite, PostgreSQL, MySQL, etc.)
+Or push a `WHERE` / `LIMIT` down to a named table with a `ReadQuery` value:
-## Configuration Options
+```haskell
+D.toMarkdown' <$> readTableWith "./data/chinook.db" "artists" (allRows & limit 3)
+```
+>
+> | ArtistId
Int | Name
Maybe Text |
+> | ----------------|------------------- |
+> | 1 | Just "AC/DC" |
+> | 2 | Just "Accept" |
+> | 3 | Just "Aerosmith" |
+
+## Tier 1: typed schema
+
+`declareTable` reads the schema at compile time and emits just the schema type. You read into it
+with `readTableTyped`, where the schema is a type argument and the database and table are ordinary
+values. Column references go through `col @"Name"`, checked against the schema, so a typo or a wrong
+type is a compile error. Nothing is keyed on a generated function name.
+
```haskell
-data PersistentConfig = PersistentConfig
- { pcBatchSize :: Int -- Number of records to fetch at once (default: 10000)
- , pcIncludeId :: Bool -- Whether to include entity ID as column (default: True)
- , pcIdColumnName :: Text -- Name for the ID column (default: "id")
- }
+import qualified DataFrame.Typed as DT
+import DataFrame.Typed ((.==.))
+import DataFrame.IO.Persistent.Schema (declareTable)
```
-## Advanced Usage
+>
-### Custom Field Extraction
+The splice brings one thing into scope, the type
+`type ArtistsSchema = '[Column "ArtistId" Int, Column "Name" (Maybe Text)]`:
-You can also extract fields from individual entities:
+```haskell
+$(declareTable "./data/chinook.db" "artists")
+```
+>
+
+`readTableTyped @ArtistsSchema` reads any database/table into a `TypedDataFrame ArtistsSchema` (it
+validates the schema as it reads). You can bind your own reader:
+
```haskell
-let user = TestUser "Alice" 25 True
- columns = persistFieldsToColumns user
--- Result: [("name", SomeColumn ["Alice"]), ("age", SomeColumn [25]), ("active", SomeColumn [True])]
+artists = readTableTyped @ArtistsSchema "./data/chinook.db" "artists"
```
-### Working with Vector Data
+>
+`thaw` drops a typed frame back to an untyped `DataFrame`:
+
```haskell
--- Extract specific column data
-let names = V.toList $ DF.columnAsVector test_user_name df
- ages = V.toList $ DF.columnAsVector test_user_age df
- activeFlags = V.toList $ DF.columnAsVector test_user_active df
+D.toMarkdown' . D.take 5 . DT.thaw <$> artists
```
-## Examples
+>
+> | ArtistId
Int | Name
Maybe Text |
+> | ----------------|------------------------- |
+> | 1 | Just "AC/DC" |
+> | 2 | Just "Accept" |
+> | 3 | Just "Aerosmith" |
+> | 4 | Just "Alanis Morissette" |
+> | 5 | Just "Alice In Chains" |
-For comprehensive examples and test cases, see:
-- [`tests/PersistentTests.hs`](tests/PersistentTests.hs) - Full test suite with examples
-- [`../docs/persistent_integration.md`](../docs/persistent_integration.md) - Detailed integration guide
+Column access is checked against the schema. `col @"Name"` only compiles because `"Name"` is a
+column of `ArtistsSchema` (its element type is `Maybe Text`):
-## Status
+```haskell
+DT.columnAsList @"Name" . DT.take 3 <$> artists
+```
-This package is **actively maintained** and tested. Current test coverage includes:
-- ✅ Entity loading with and without filters
-- ✅ Custom configuration options
-- ✅ DataFrame operations on Persistent data
-- ✅ Empty result set handling
-- ✅ Field extraction utilities
-- ✅ Multi-table relationships
+>
+> [Just "AC/DC",Just "Accept",Just "Aerosmith"]
-## Documentation
+A filter on a column that doesn't exist (or has the wrong type) is a compile error rather than a
+runtime surprise:
-For detailed documentation, see:
-- [Main dataframe documentation](https://github.com/mchav/dataframe)
-- [Persistent integration guide](../docs/persistent_integration.md)
+```haskell
+D.toMarkdown' . DT.thaw . DT.filterWhere (DT.col @"Name" .==. DT.lit (Just "Accept")) <$> artists
+```
-## License
+>
+> | ArtistId
Int | Name
Maybe Text |
+> | ----------------|------------------- |
+> | 2 | Just "Accept" |
-GPL-3.0-or-later (same as the main dataframe package)
+Because the database is a value, reading the same table from two sources to join them is just two
+calls with the same `@ArtistsSchema`:
+
+```text
+a <- readTableTyped @ArtistsSchema "europe.sqlite" "artists"
+b <- readTableTyped @ArtistsSchema "us.sqlite" "artists"
+-- DT.thaw a / DT.thaw b, then DataFrame.innerJoin on "ArtistId", etc.
+```
+
+(`readSqlTyped @cols db "SELECT ... JOIN ..."` does the same for an arbitrary query. The Postgres
+section below reads this same `ArtistsSchema` from a different backend.)
+
+## Tier 2: generate a `persistent` entity
+
+If you want the full `persistent` experience (the typed `Filter` DSL, relations, write-back),
+`declareEntity` builds the entity from the live schema, so you skip the `persistLowerCase` block. It
+needs the usual `persistent` extensions and unqualified `Text` / `Int64` in scope:
+
+```haskell
+-- cabal: build-depends: persistent, persistent-sqlite
+-- cabal: default-extensions: GADTs, TypeFamilies, FlexibleInstances, MultiParamTypeClasses
+-- cabal: default-extensions: StandaloneDeriving, DerivingStrategies, GeneralizedNewtypeDeriving
+-- cabal: default-extensions: UndecidableInstances
+import Data.Int (Int64)
+import Data.Text (Text)
+import Database.Persist (Filter, SelectOpt (..), (<-.), (==.))
+import Database.Persist.Sqlite (runSqlite)
+import DataFrame.IO.Persistent.Schema (declareEntity)
+```
+
+>
+
+This generates the `Albums` entity (`Id`, `title`, `artistId`), bound to the real
+`albums` / `AlbumId` / `Title` / `ArtistId` SQLite names, plus its `EntityField` constructors:
+
+```haskell
+$(declareEntity "./data/chinook.db" "albums")
+```
+
+>
+
+Now load it with `selectToDataFrame`. It's a generic loader: it works for any `persistent` entity,
+so the one `declareEntity` splice is all the boilerplate.
+
+```text
+selectToDataFrame
+ :: (MonadIO m, PersistEntity r, PersistEntityBackend r ~ SqlBackend)
+ => [Filter r] -- persistent's typed WHERE DSL, e.g. [AlbumsArtistId ==. 1]
+ -> [SelectOpt r] -- ordering / paging, e.g. [Asc AlbumsTitle, LimitTo 10]
+ -> ReaderT SqlBackend m DataFrame
+```
+
+It runs persistent's `selectList filters opts` and turns the `[Entity Albums]` into a `DataFrame`:
+the entity key becomes an `id` column and each field becomes a column (named by its
+Haskell field, e.g. `title`, `artistId`), with element types decoded from the stored values. The
+`[Filter Albums]` and `[SelectOpt Albums]` you pass are persistent's own, checked against the
+entity, so a filter for the wrong field won't compile.
+
+Find AC/DC's albums (`ArtistId` 1):
+
+```haskell
+D.toMarkdown' <$> runSqlite "./data/chinook.db" (selectToDataFrame [AlbumsArtistId ==. 1] [])
+```
+
+>
+> | id
Int | title
Text | artistId
Int |
+> | ----------|---------------------------------------|---------------- |
+> | 1 | For Those About To Rock We Salute You | 1 |
+> | 4 | Let There Be Rock | 1 |
+
+You can use both arguments together: filter and order/page in one query. Here are the albums by
+AC/DC or Alanis Morissette (`ArtistId` 1 or 4), sorted by title, capped at 5. (`<-.` is persistent's
+"field in list"; an empty filter list loads everything.)
+
+```haskell
+D.toMarkdown' <$> runSqlite "./data/chinook.db"
+ (selectToDataFrame [AlbumsArtistId <-. [1, 4]] [Asc AlbumsTitle, LimitTo 5])
+```
+
+>
+> | id
Int | title
Text | artistId
Int |
+> | ----------|---------------------------------------|---------------- |
+> | 1 | For Those About To Rock We Salute You | 1 |
+> | 6 | Jagged Little Pill | 4 |
+> | 4 | Let There Be Rock | 1 |
+
+## Hand-off between dataframe and persistent
+
+The `...Conn` readers run on a `ReaderT SqlBackend m`, so a raw read, an entity read, and ordinary
+`persistent` queries all compose in one `runSqlite` transaction (this works on Postgres/MySQL too).
+Here both reads share one connection and return each frame's `(rows, columns)`:
+
+```haskell
+runSqlite "./data/chinook.db" $ do
+ artists <- readTableConn "artists" -- raw → DataFrame
+ albums <- selectToDataFrame ([] :: [Filter Albums]) [] -- entity → DataFrame
+ pure (D.dimensions artists, D.dimensions albums)
+```
+
+>
+> ((275,2),(347,3))
+
+## PostgreSQL (and other backends)
+
+Everything except the file-path helpers and the compile-time splices is backend-agnostic: the
+`...Conn` readers and `selectToDataFrame` run on any `persistent` `SqlBackend`. Introspection picks
+`PRAGMA` for SQLite and `information_schema` for PostgreSQL/MySQL. The library doesn't depend on
+`persistent-postgresql`; you bring the connection.
+
+These examples run against a real PostgreSQL with the same `artists` / `albums` data, and the
+notebook starts it itself. The block below uses GHCi's `:!` shell escape to spin up a throwaway
+cluster on port 54329 and load the two tables from the SQLite fixture. It needs `postgres` and
+`sqlite3` on `PATH` (e.g. `brew install postgresql@16`). A matching teardown runs at the end of the
+section.
+
+```haskell
+:! bash scripts/pg-setup.sh
+```
+
+>
+> postgres ready on port 54329 (db chinook: artists + albums)
+
+Open a connection (you supply `persistent-postgresql`; `runPg` is the usual `runSqlConn` wrapper):
+
+```haskell
+-- cabal: build-depends: persistent, persistent-postgresql, transformers, resourcet, monad-logger
+import Control.Monad.Trans.Reader (ReaderT)
+import Control.Monad.Trans.Resource (ResourceT, runResourceT)
+import Control.Monad.Logger (NoLoggingT, runNoLoggingT)
+import Database.Persist.Sql (SqlBackend)
+import Database.Persist.Postgresql (runSqlConn, withPostgresqlConn)
+
+runPg :: ReaderT SqlBackend (ResourceT (NoLoggingT IO)) a -> IO a
+runPg act =
+ runNoLoggingT (runResourceT (withPostgresqlConn "host=localhost port=54329 dbname=chinook user=postgres" (runSqlConn act)))
+```
+
+>
+
+Discovery and reads use the same functions as SQLite, just `...Conn` wrapped in `runPg`:
+
+```haskell
+runPg listTablesConn
+```
+
+>
+> ["albums","artists"]
+
+```haskell
+D.toMarkdown' <$> runPg (describeTableConn "artists")
+```
+
+>
+> | Column Name
Text | Type
Text | SQLite Type
Text | Nullable
Bool | Primary Key
Bool |
+> | --------------------|--------------|---------------------|------------------|-------------------- |
+> | ArtistId | Int | integer | False | True |
+> | Name | Maybe Text | text | True | False |
+
+```haskell
+D.toMarkdown' . D.take 5 <$> runPg (readTableConn "artists")
+```
+
+>
+> | ArtistId
Int | Name
Maybe Text |
+> | ----------------|------------------------- |
+> | 1 | Just "AC/DC" |
+> | 2 | Just "Accept" |
+> | 3 | Just "Aerosmith" |
+> | 4 | Just "Alanis Morissette" |
+> | 5 | Just "Alice In Chains" |
+
+The typed reader is backend-agnostic too. The same `ArtistsSchema` generated from the SQLite file in
+Tier 1 validates this PostgreSQL read, so it's one schema type across two databases:
+
+```haskell
+D.toMarkdown' . D.take 3 . DT.thaw <$> runPg (readTableTypedConn @ArtistsSchema "artists")
+```
+
+>
+> | ArtistId
Int | Name
Maybe Text |
+> | ----------------|------------------- |
+> | 1 | Just "AC/DC" |
+> | 2 | Just "Accept" |
+> | 3 | Just "Aerosmith" |
+
+The `Albums` entity generated from SQLite at compile time is a plain `persistent` entity, so the
+same `declareEntity` splice and `selectToDataFrame` run unchanged against PostgreSQL:
+
+```haskell
+D.toMarkdown' <$> runPg (selectToDataFrame [AlbumsArtistId <-. [1, 4]] [Asc AlbumsTitle, LimitTo 5])
+```
+
+>
+> | id
Int | title
Text | artistId
Int |
+> | ----------|---------------------------------------|---------------- |
+> | 1 | For Those About To Rock We Salute You | 1 |
+> | 6 | Jagged Little Pill | 4 |
+> | 4 | Let There Be Rock | 1 |
+
+Tear the throwaway cluster back down (again via `:!`):
+
+```haskell
+:! bash scripts/pg-teardown.sh
+```
+
+>
+> postgres stopped and removed
+
+## How types are inferred
+
+SQLite type affinities map to Haskell element types as below. `NOT NULL` columns are non-null,
+everything else becomes `Maybe`. You can override a column with `typeOverride` in `DeclareOptions`
+(`declareTableWith` / `declareEntityWith`).
+
+| SQLite declared type | Haskell type |
+|----------------------|--------------|
+| `INTEGER`, `INT...` | `Int` |
+| `REAL`, `FLOAT`, `DOUBLE`, `NUMERIC`, `DECIMAL` | `Double` |
+| `TEXT`, `VARCHAR`, `CHAR`, `CLOB` | `Text` |
+| `BOOLEAN` | `Bool` |
+| `BLOB` / no declared type | `ByteString` |
+| `DATE` | `Day` |
+| `DATETIME`, `TIMESTAMP` | `UTCTime` |
+| `TIME` | `TimeOfDay` |
+
+## Install
+
+```
+build-depends: dataframe, dataframe-persistent, persistent-sqlite
+```
+
+The classic path still works and is unchanged: hand-write a `persistent` entity with
+`share`/`persistLowerCase` and `derivePersistentDataFrame`, then load with `fromPersistent`.
diff --git a/dataframe-persistent.cabal b/dataframe-persistent.cabal
--- a/dataframe-persistent.cabal
+++ b/dataframe-persistent.cabal
@@ -1,13 +1,13 @@
cabal-version: 2.4
name: dataframe-persistent
-version: 0.2.0.0
+version: 0.3.0.0
synopsis: Persistent database integration for the dataframe library
description: This package provides integration between the dataframe library and the Persistent database library, allowing you to load database entities into DataFrames and save DataFrames back to the database.
bug-reports: https://github.com/mchav/dataframe/issues
-license: GPL-3.0-or-later
+license: MIT
license-file: ./LICENSE
author: Michael Chavinda, Junji Hashimoto
maintainer: mschavinda@gmail.com
@@ -15,7 +15,7 @@
copyright: (c) 2024-2025 Michael Chavinda
category: Data, Database
tested-with: GHC ==9.4.8 || ==9.6.7 || ==9.8.4 || ==9.10.3 || ==9.12.2
-extra-doc-files: ./CHANGELOG.md ./README.md
+extra-doc-files: CHANGELOG.md README.md
source-repository head
type: git
@@ -33,12 +33,22 @@
import: warnings
default-extensions: Strict
exposed-modules: DataFrame.IO.Persistent,
- DataFrame.IO.Persistent.TH
+ DataFrame.IO.Persistent.TH,
+ DataFrame.IO.Persistent.Read,
+ DataFrame.IO.Persistent.Read.Columns,
+ DataFrame.IO.Persistent.Read.Sqlite,
+ DataFrame.IO.Persistent.Schema,
+ DataFrame.IO.Persistent.Schema.Common,
+ DataFrame.IO.Persistent.Schema.Introspect
build-depends: base >= 4 && <5,
- bytestring >= 0.11 && < 0.13,
+ bytestring >= 0.10 && < 0.13,
+ conduit >= 1.3 && < 1.4,
containers >= 0.6.7 && < 0.9,
- dataframe ^>= 0.4,
+ dataframe-core ^>= 1.0,
+ dataframe-operations ^>= 1.0,
persistent >= 2.14 && < 3,
+ persistent-sqlite >= 2.13 && < 3,
+ resourcet >= 1.2 && < 2,
template-haskell >= 2.0 && < 3,
text >= 2.0 && < 3,
time >= 1.12 && < 2,
@@ -51,9 +61,12 @@
import: warnings
type: exitcode-stdio-1.0
main-is: Main.hs
- other-modules: PersistentTests
+ other-modules: PersistentTests,
+ SchemaTHTests
build-depends: base >= 4 && < 5,
- dataframe ^>= 0.4,
+ dataframe >= 1 && < 3,
+ dataframe-core ^>= 1.0,
+ dataframe-operations ^>= 1.0,
dataframe-persistent,
HUnit ^>= 1.6,
monad-logger >= 0.3 && < 0.4,
diff --git a/src/DataFrame/IO/Persistent.hs b/src/DataFrame/IO/Persistent.hs
--- a/src/DataFrame/IO/Persistent.hs
+++ b/src/DataFrame/IO/Persistent.hs
@@ -1,5 +1,4 @@
{-# LANGUAGE AllowAmbiguousTypes #-}
-{-# LANGUAGE BangPatterns #-}
{-# LANGUAGE ConstraintKinds #-}
{-# LANGUAGE DataKinds #-}
{-# LANGUAGE ExistentialQuantification #-}
@@ -42,19 +41,16 @@
import Control.Monad (forM)
import Control.Monad.IO.Class (MonadIO, liftIO)
import Control.Monad.Trans.Reader (ReaderT)
-import Data.ByteString (ByteString)
import qualified Data.Map.Strict as M
import Data.Proxy (Proxy (..))
import Data.Text (Text)
import qualified Data.Text as T
-import Data.Time (Day, TimeOfDay, UTCTime)
import qualified Data.Vector as V
import qualified DataFrame.Internal.Column as DFCol
import DataFrame.Internal.DataFrame (DataFrame (..))
import qualified DataFrame.Internal.DataFrame as DF
import Database.Persist
import Database.Persist.Sql hiding (Column)
-import Database.Persist.Types (fieldHaskell, getEntityFields, unFieldNameHS)
import Unsafe.Coerce (unsafeCoerce)
-- | Get number of rows in a DataFrame
@@ -147,7 +143,7 @@
numCols = length colNames
dimensions = (0, numCols)
emptyColumns = replicate numCols (DFCol.fromList ([] :: [Text]))
- return $ DataFrame (V.fromList emptyColumns) indices dimensions
+ return $ DataFrame (V.fromList emptyColumns) indices dimensions M.empty
else do
-- Create columns from entity data
columns <- createColumnsFromEntities @record config entities
@@ -155,7 +151,7 @@
numRows = if null entities then 0 else length entities
numCols = length colNames
dimensions = (numRows, numCols)
- return $ DataFrame (V.fromList columns) indices dimensions
+ return $ DataFrame (V.fromList columns) indices dimensions M.empty
-- | Create columns from entities
createColumnsFromEntities ::
@@ -217,11 +213,10 @@
ReaderT SqlBackend m [Key record]
toPersistent df = do
let rowCount = nRows df
- keys <- forM [0 .. rowCount - 1] $ \i -> do
+ forM [0 .. rowCount - 1] $ \i -> do
case rowToEntity i df of
Left err -> error $ "Failed to convert row " <> show i <> ": " <> err
Right entity -> insert (entityVal entity)
- return keys
-- | Convert entity fields to columns (helper for implementations)
entityToColumns ::
diff --git a/src/DataFrame/IO/Persistent/Read.hs b/src/DataFrame/IO/Persistent/Read.hs
new file mode 100644
--- /dev/null
+++ b/src/DataFrame/IO/Persistent/Read.hs
@@ -0,0 +1,291 @@
+{-# LANGUAGE AllowAmbiguousTypes #-}
+{-# LANGUAGE FlexibleContexts #-}
+{-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE ScopedTypeVariables #-}
+{-# LANGUAGE TypeApplications #-}
+{-# LANGUAGE TypeFamilies #-}
+{-# LANGUAGE TypeOperators #-}
+
+{- |
+Module : DataFrame.IO.Persistent.Read
+License : MIT
+
+The runtime front door for loading SQLite into a 'DataFrame'. Mirrors the
+ergonomics of @readCsv@: point at a file and a table and get a 'DataFrame', with
+column types inferred from the SQLite schema. No Template Haskell, no entity
+boilerplate, and it works directly in GHCi.
+
+For compile-time typed schemas use "DataFrame.IO.Persistent.Schema"; for the
+@persistent@ entity path use "DataFrame.IO.Persistent".
+-}
+module DataFrame.IO.Persistent.Read (
+ -- * Reading a whole table (types from the schema)
+ readTable,
+ readTableWith,
+ readTableConn,
+ readTableConnWith,
+
+ -- * Reading an arbitrary query (types sniffed from values)
+ readSql,
+ readSqlWith,
+
+ -- * Reading straight into a typed schema (schema as a type argument)
+ readTableTyped,
+ readTableTypedConn,
+ readSqlTyped,
+
+ -- * Loading @persistent@ entities generically (no instance required)
+ selectToDataFrame,
+
+ -- * Discovery
+ listTables,
+ listTablesConn,
+ describeTable,
+ describeTableConn,
+
+ -- * Filtering pushed down to SQLite
+ ReadQuery (..),
+ allRows,
+ where_,
+ limit,
+ orderBy,
+
+ -- * Column builders (used by generated readers)
+ ColumnReader,
+ buildColumn,
+ buildNullableColumn,
+ inferColumn,
+ freezeOrThrow,
+
+ -- * Re-exports
+ ColumnInfo (..),
+) where
+
+import Control.Monad.IO.Class (MonadIO)
+import Control.Monad.Trans.Reader (ReaderT)
+import Control.Monad.Trans.Resource (MonadResource)
+import Data.List (transpose)
+import Data.Text (Text)
+import qualified Data.Text as T
+import Database.Persist
+import Database.Persist.Sql (SqlBackend)
+import Database.Persist.Sqlite (runSqlite)
+
+import DataFrame.IO.Persistent.Read.Columns
+import DataFrame.IO.Persistent.Read.Sqlite (runNamedQuery)
+import DataFrame.IO.Persistent.Schema.Introspect
+import DataFrame.Internal.Column (fromList)
+import DataFrame.Internal.DataFrame (DataFrame, fromNamedColumns)
+import DataFrame.Typed.Schema (KnownSchema)
+import DataFrame.Typed.Types (TypedDataFrame)
+
+{- | A SQLite-side filter pushed into the generated @SELECT@. Build with
+'allRows' and the '&'-friendly combinators 'where_', 'limit', 'orderBy'.
+-}
+data ReadQuery = ReadQuery
+ { rqWhere :: Maybe Text
+ -- ^ A @WHERE@ body (without the @WHERE@ keyword); may contain @?@ placeholders.
+ , rqParams :: [PersistValue]
+ -- ^ Values bound to the placeholders in 'rqWhere'.
+ , rqOrderBy :: Maybe Text
+ -- ^ An @ORDER BY@ body (without the keyword).
+ , rqLimit :: Maybe Int
+ , rqOffset :: Maybe Int
+ }
+ deriving (Eq, Show)
+
+-- | The unfiltered query: every row, every column.
+allRows :: ReadQuery
+allRows = ReadQuery Nothing [] Nothing Nothing Nothing
+
+-- | Add a @WHERE@ clause (with bound parameters) to a 'ReadQuery'.
+where_ :: Text -> [PersistValue] -> ReadQuery -> ReadQuery
+where_ clause params q = q{rqWhere = Just clause, rqParams = params}
+
+-- | Limit the number of rows fetched.
+limit :: Int -> ReadQuery -> ReadQuery
+limit n q = q{rqLimit = Just n}
+
+-- | Add an @ORDER BY@ clause.
+orderBy :: Text -> ReadQuery -> ReadQuery
+orderBy clause q = q{rqOrderBy = Just clause}
+
+{- | Read an entire table into a 'DataFrame', inferring column types (and
+nullability) from the SQLite schema. The @pd.read_sql_table@ analogue.
+-}
+readTable :: FilePath -> Text -> IO DataFrame
+readTable path table = readTableWith path table allRows
+
+-- | 'readTable' with a SQLite-side filter (e.g. for tables larger than memory).
+readTableWith :: FilePath -> Text -> ReadQuery -> IO DataFrame
+readTableWith path table q = do
+ readers <- columnReadersFor path table
+ runSqlite (T.pack path) (readColumnsConn table readers q)
+
+{- | 'readTable' over an existing @persistent@ connection (any 'SqlBackend',
+so it shares a transaction with @fromPersistent@ / @insert@).
+-}
+readTableConn :: (MonadResource m) => Text -> ReaderT SqlBackend m DataFrame
+readTableConn table = readTableConnWith table allRows
+
+-- | 'readTableConn' with a SQLite-side filter.
+readTableConnWith ::
+ (MonadResource m) => Text -> ReadQuery -> ReaderT SqlBackend m DataFrame
+readTableConnWith table q = do
+ cols <- introspectTableConn table
+ readColumnsConn table (map columnInfoReader cols) q
+
+columnReadersFor :: FilePath -> Text -> IO [(Text, ColumnReader)]
+columnReadersFor path table =
+ map columnInfoReader <$> introspectTable path (T.unpack table)
+
+columnInfoReader :: ColumnInfo -> (Text, ColumnReader)
+columnInfoReader ci =
+ (ciName ci, columnReaderFor (inferType (ciDeclType ci)) (not (ciNotNull ci)))
+
+{- | Run an arbitrary SQL query into a 'DataFrame'. Result columns keep their
+query names (aliases included); types are inferred from the returned values.
+-}
+readSql :: FilePath -> Text -> IO DataFrame
+readSql path sql = readSqlWith path sql []
+
+-- | 'readSql' with bound query parameters (injection-safe).
+readSqlWith :: FilePath -> Text -> [PersistValue] -> IO DataFrame
+readSqlWith path sql params = do
+ (names, rows) <- runNamedQuery (T.pack path) sql params
+ pure (assembleInferred names rows)
+
+{- | Read a whole table straight into a typed 'TypedDataFrame', validating the
+schema at the boundary. The schema is a /type argument/ while the database and
+table are ordinary values, so the same table name in two different databases is
+just two calls with the same @\@Schema@, no generated names to collide:
+
+@
+a <- readTableTyped \@ArtistsSchema "db1.sqlite" "artists"
+b <- readTableTyped \@ArtistsSchema "db2.sqlite" "artists"
+-- ... then join a and b
+@
+-}
+readTableTyped ::
+ forall cols. (KnownSchema cols) => FilePath -> Text -> IO (TypedDataFrame cols)
+readTableTyped path table = readTable path table >>= freezeOrThrow
+
+-- | 'readTableTyped' over an existing connection (any backend).
+readTableTypedConn ::
+ forall cols m.
+ (KnownSchema cols, MonadResource m) =>
+ Text -> ReaderT SqlBackend m (TypedDataFrame cols)
+readTableTypedConn table = readTableConn table >>= freezeOrThrow
+
+-- | Read an arbitrary query (e.g. a join) straight into a typed 'TypedDataFrame'.
+readSqlTyped ::
+ forall cols. (KnownSchema cols) => FilePath -> Text -> IO (TypedDataFrame cols)
+readSqlTyped path sql = readSql path sql >>= freezeOrThrow
+
+-- | List the user tables in a SQLite database.
+listTables :: FilePath -> IO [Text]
+listTables = introspectTableNames
+
+{- | 'listTables' over an existing connection (works on any backend, e.g. a
+PostgreSQL 'SqlBackend' opened with @withPostgresqlConn@).
+-}
+listTablesConn :: (MonadResource m) => ReaderT SqlBackend m [Text]
+listTablesConn = introspectTableNamesConn
+
+{- | A @describeColumns@-style summary of a table's schema (one row per column),
+read from the schema without scanning the data.
+-}
+describeTable :: FilePath -> Text -> IO DataFrame
+describeTable path table = schemaFrame <$> introspectTable path (T.unpack table)
+
+-- | 'describeTable' over an existing connection (any backend).
+describeTableConn :: (MonadResource m) => Text -> ReaderT SqlBackend m DataFrame
+describeTableConn table = schemaFrame <$> introspectTableConn table
+
+schemaFrame :: [ColumnInfo] -> DataFrame
+schemaFrame cols =
+ fromNamedColumns
+ [ ("Column Name", fromList (map ciName cols))
+ , ("Type", fromList (map columnHaskellType cols))
+ , ("SQLite Type", fromList (map ciDeclType cols))
+ , ("Nullable", fromList (map (not . ciNotNull) cols))
+ , ("Primary Key", fromList (map ciPk cols))
+ ]
+
+columnHaskellType :: ColumnInfo -> Text
+columnHaskellType c =
+ let base = haskellTypeName (inferType (ciDeclType c))
+ in if ciNotNull c then base else "Maybe " <> base
+
+{- | Load any @persistent@ entity into a 'DataFrame' without deriving an
+instance: filters give you the typed @persistent@ DSL, the rows are decoded
+generically. Powers the @declareEntity@ workflow.
+-}
+selectToDataFrame ::
+ forall r m.
+ ( MonadIO m
+ , PersistEntity r
+ , PersistEntityBackend r ~ SqlBackend
+ ) =>
+ [Filter r] -> [SelectOpt r] -> ReaderT SqlBackend m DataFrame
+selectToDataFrame filters opts =
+ entitiesToDataFrame <$> selectList filters opts
+
+entitiesToDataFrame :: forall r. (PersistEntity r) => [Entity r] -> DataFrame
+entitiesToDataFrame ents =
+ assembleInferred ("id" : entityFieldNames @r) (map entityRow ents)
+
+-- @keyToValues@ decodes any key (single or backend) without a @ToBackendKey@
+-- constraint; single-column keys (the common case) yield the @id@ column.
+entityRow :: (PersistEntity r) => Entity r -> [PersistValue]
+entityRow (Entity k v) = keyToValues k ++ toPersistFields v
+
+entityFieldNames :: forall r. (PersistEntity r) => [Text]
+entityFieldNames =
+ map
+ (unFieldNameHS . fieldHaskell)
+ (getEntityFields (entityDef (Nothing :: Maybe r)))
+
+{- | The streaming engine behind every table reader: build the @SELECT@, fetch
+rows, transpose into columns, and apply each column's reader.
+-}
+readColumnsConn ::
+ (MonadResource m) =>
+ Text -> [(Text, ColumnReader)] -> ReadQuery -> ReaderT SqlBackend m DataFrame
+readColumnsConn table cols q = do
+ rows <- runRawRows (selectSql table (map fst cols) q) (rqParams q)
+ pure (assembleColumns cols rows)
+
+assembleColumns :: [(Text, ColumnReader)] -> [[PersistValue]] -> DataFrame
+assembleColumns cols rows =
+ fromNamedColumns (zipWith apply cols (cells (length cols) rows))
+ where
+ apply (name, rdr) pvs = (name, rdr pvs)
+
+assembleInferred :: [Text] -> [[PersistValue]] -> DataFrame
+assembleInferred names rows =
+ fromNamedColumns (zip names (map inferColumn (cells (length names) rows)))
+
+{- | Transpose rows into per-column value lists, padding to @n@ empty columns
+when there are no rows (so column structure survives an empty result).
+-}
+cells :: Int -> [[PersistValue]] -> [[PersistValue]]
+cells n [] = replicate n []
+cells _ rows = transpose rows
+
+selectSql :: Text -> [Text] -> ReadQuery -> Text
+selectSql table cols q =
+ T.unwords (filter (not . T.null) clauses)
+ where
+ colList = if null cols then "*" else T.intercalate ", " (map quoteIdent cols)
+ clauses =
+ [ "SELECT " <> colList
+ , "FROM " <> quoteIdent table
+ , maybe "" ("WHERE " <>) (rqWhere q)
+ , maybe "" ("ORDER BY " <>) (rqOrderBy q)
+ , maybe "" (("LIMIT " <>) . tshow) (rqLimit q)
+ , maybe "" (("OFFSET " <>) . tshow) (rqOffset q)
+ ]
+
+tshow :: (Show a) => a -> Text
+tshow = T.pack . show
diff --git a/src/DataFrame/IO/Persistent/Read/Columns.hs b/src/DataFrame/IO/Persistent/Read/Columns.hs
new file mode 100644
--- /dev/null
+++ b/src/DataFrame/IO/Persistent/Read/Columns.hs
@@ -0,0 +1,121 @@
+{-# LANGUAGE AllowAmbiguousTypes #-}
+{-# LANGUAGE FlexibleContexts #-}
+{-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE ScopedTypeVariables #-}
+{-# LANGUAGE TypeApplications #-}
+-- 'buildNullableColumn' deliberately constrains @Columnable (Maybe a)@ (needed to
+-- build a bitmap-backed nullable column); GHC flags that as simplifiable.
+{-# OPTIONS_GHC -Wno-simplifiable-class-constraints #-}
+
+{- |
+Module : DataFrame.IO.Persistent.Read.Columns
+License : MIT
+
+Turning rows of 'PersistValue's into typed dataframe 'Column's: decoders driven
+by a known 'HaskellType', value-sniffing inference for unknown columns, and the
+typed-schema boundary. Shared by the runtime readers and the generated readers.
+-}
+module DataFrame.IO.Persistent.Read.Columns (
+ ColumnReader,
+ buildColumn,
+ buildNullableColumn,
+ columnReaderFor,
+ inferColumn,
+ freezeOrThrow,
+) where
+
+import Control.Monad.IO.Class (MonadIO, liftIO)
+import Data.ByteString (ByteString)
+import Data.List (find)
+import Data.Text (Text)
+import qualified Data.Text as T
+import Data.Time (Day, TimeOfDay, UTCTime)
+import Data.Typeable (Proxy (..), typeRep)
+import Database.Persist (PersistField (fromPersistValue), PersistValue (..))
+
+import DataFrame.IO.Persistent.Schema.Introspect (HaskellType (..))
+import DataFrame.Internal.Column (Column, Columnable, fromList)
+import DataFrame.Internal.DataFrame (DataFrame)
+import DataFrame.Typed.Freeze (freezeWithError)
+import DataFrame.Typed.Schema (KnownSchema)
+import DataFrame.Typed.Types (TypedDataFrame)
+
+-- | A function that turns one column's worth of 'PersistValue's into a 'Column'.
+type ColumnReader = [PersistValue] -> Column
+
+-- | Build a non-null column, decoding each value with 'fromPersistValue'.
+buildColumn :: forall a. (Columnable a, PersistField a) => ColumnReader
+buildColumn pvs = case traverse fromPersistValue pvs of
+ Left err -> error (decodeError (typeName @a) err)
+ Right (xs :: [a]) -> fromList xs
+
+-- | Build a nullable column (@PersistNull@ becomes @Nothing@).
+buildNullableColumn ::
+ forall a. (Columnable (Maybe a), PersistField a) => ColumnReader
+buildNullableColumn pvs = case traverse fromPersistValue pvs of
+ Left err -> error (decodeError (typeName @(Maybe a)) err)
+ Right (xs :: [Maybe a]) -> fromList xs
+
+{- | Infer a column's type by sniffing the first non-null value; nullable if any
+value is @PersistNull@. Used for arbitrary queries and generic entity loads.
+-}
+inferColumn :: [PersistValue] -> Column
+inferColumn pvs = columnReaderFor (inferHaskellType pvs) (any isNull pvs) pvs
+
+columnReaderFor :: HaskellType -> Bool -> ColumnReader
+columnReaderFor ht True = nullableReaderFor ht
+columnReaderFor ht False = readerFor ht
+
+readerFor :: HaskellType -> ColumnReader
+readerFor HTInt = buildColumn @Int
+readerFor HTDouble = buildColumn @Double
+readerFor HTText = buildColumn @Text
+readerFor HTBool = buildColumn @Bool
+readerFor HTByteString = buildColumn @ByteString
+readerFor HTDay = buildColumn @Day
+readerFor HTUTCTime = buildColumn @UTCTime
+readerFor HTTimeOfDay = buildColumn @TimeOfDay
+
+nullableReaderFor :: HaskellType -> ColumnReader
+nullableReaderFor HTInt = buildNullableColumn @Int
+nullableReaderFor HTDouble = buildNullableColumn @Double
+nullableReaderFor HTText = buildNullableColumn @Text
+nullableReaderFor HTBool = buildNullableColumn @Bool
+nullableReaderFor HTByteString = buildNullableColumn @ByteString
+nullableReaderFor HTDay = buildNullableColumn @Day
+nullableReaderFor HTUTCTime = buildNullableColumn @UTCTime
+nullableReaderFor HTTimeOfDay = buildNullableColumn @TimeOfDay
+
+inferHaskellType :: [PersistValue] -> HaskellType
+inferHaskellType = maybe HTText haskellTypeOf . find (not . isNull)
+
+haskellTypeOf :: PersistValue -> HaskellType
+haskellTypeOf (PersistInt64 _) = HTInt
+haskellTypeOf (PersistDouble _) = HTDouble
+haskellTypeOf (PersistBool _) = HTBool
+haskellTypeOf (PersistByteString _) = HTByteString
+haskellTypeOf (PersistDay _) = HTDay
+haskellTypeOf (PersistUTCTime _) = HTUTCTime
+haskellTypeOf (PersistTimeOfDay _) = HTTimeOfDay
+haskellTypeOf _ = HTText
+
+isNull :: PersistValue -> Bool
+isNull PersistNull = True
+isNull _ = False
+
+typeName :: forall a. (Columnable a) => String
+typeName = show (typeRep (Proxy @a))
+
+decodeError :: String -> Text -> String
+decodeError ty err = "buildColumn: failed to decode column as " <> ty <> ": " <> T.unpack err
+
+{- | Validate an untyped 'DataFrame' against a typed schema and wrap it, throwing
+an 'IOError' on mismatch. The generated @declareTable@ readers call this; the
+schema came from the same database, so success is the normal case.
+-}
+freezeOrThrow ::
+ forall cols m.
+ (KnownSchema cols, MonadIO m) =>
+ DataFrame -> m (TypedDataFrame cols)
+freezeOrThrow df =
+ either (liftIO . ioError . userError . T.unpack) pure (freezeWithError @cols df)
diff --git a/src/DataFrame/IO/Persistent/Read/Sqlite.hs b/src/DataFrame/IO/Persistent/Read/Sqlite.hs
new file mode 100644
--- /dev/null
+++ b/src/DataFrame/IO/Persistent/Read/Sqlite.hs
@@ -0,0 +1,65 @@
+{- |
+Module : DataFrame.IO.Persistent.Read.Sqlite
+License : MIT
+
+A minimal direct-SQLite query that returns result column /names/ alongside the
+values. @persistent@'s @rawQuery@ drops the names, so for arbitrary @readSql@
+queries we prepare the statement ourselves and read @sqlite3_column_name@, which
+works even for aliased expressions and empty result sets.
+-}
+module DataFrame.IO.Persistent.Read.Sqlite (
+ runNamedQuery,
+) where
+
+import Control.Exception (bracket)
+import qualified Data.ByteString as BS
+import Data.Text (Text)
+import qualified Data.Text.Encoding as TE
+import Database.Persist (PersistValue)
+import Database.Sqlite (
+ StepResult (Done, Row),
+ bind,
+ close,
+ columns,
+ finalize,
+ open,
+ prepare,
+ step,
+ )
+import Database.Sqlite.Internal (Statement (Statement))
+import Foreign.C.String (CString)
+import Foreign.C.Types (CInt (..))
+import Foreign.Ptr (Ptr)
+
+-- @sqlite3_column_count@ / @sqlite3_column_name@ are not re-exported by
+-- @Database.Sqlite@, so we bind them directly against the statement pointer.
+foreign import ccall unsafe "sqlite3_column_count"
+ c_sqlite3_column_count :: Ptr () -> IO CInt
+
+foreign import ccall unsafe "sqlite3_column_name"
+ c_sqlite3_column_name :: Ptr () -> CInt -> IO CString
+
+{- | Run @sql@ (with bound @params@) against the database at @dbPath@ and return
+the result column names together with every row's values.
+-}
+runNamedQuery :: Text -> Text -> [PersistValue] -> IO ([Text], [[PersistValue]])
+runNamedQuery dbPath sql params =
+ bracket (open dbPath) close $ \conn ->
+ bracket (prepare conn sql) finalize $ \stmt@(Statement p) -> do
+ bind stmt params
+ n <- fromIntegral <$> c_sqlite3_column_count p
+ names <- mapM (columnName p) [0 .. n - 1]
+ rows <- stepRows stmt
+ pure (names, rows)
+
+columnName :: Ptr () -> Int -> IO Text
+columnName p i = do
+ cstr <- c_sqlite3_column_name p (fromIntegral i)
+ TE.decodeUtf8 <$> BS.packCString cstr
+
+stepRows :: Statement -> IO [[PersistValue]]
+stepRows stmt = do
+ result <- step stmt
+ case result of
+ Row -> (:) <$> columns stmt <*> stepRows stmt
+ Done -> pure []
diff --git a/src/DataFrame/IO/Persistent/Schema.hs b/src/DataFrame/IO/Persistent/Schema.hs
new file mode 100644
--- /dev/null
+++ b/src/DataFrame/IO/Persistent/Schema.hs
@@ -0,0 +1,159 @@
+{-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE TemplateHaskellQuotes #-}
+
+{- |
+Module : DataFrame.IO.Persistent.Schema
+License : MIT
+
+Template Haskell that reads a SQLite table's schema __at compile time__ and
+generates either a typed-schema type synonym ('declareTable') or a @persistent@
+entity ('declareEntity'). Both replace the hand-written @persistLowerCase@ +
+@derivePersistentDataFrame@ boilerplate.
+
+'declareTable' emits only the schema /type/. You read into it with the generic
+'DataFrame.IO.Persistent.Read.readTableTyped', so the database and table are
+runtime values and the same schema can be reused across databases.
+
+@
+-- Typed schema, checked at compile time:
+\$('declareTable' "chinook.db" "artists")
+-- generates: type ArtistsSchema = '[Column "ArtistId" Int, Column "Name" (Maybe Text)]
+-- read any database/table into it:
+-- readTableTyped \@ArtistsSchema "chinook.db" "artists" :: IO (TypedDataFrame ArtistsSchema)
+
+-- Persistent entity: full Filter DSL, write-back, no hand-written schema.
+\$('declareEntity' "chinook.db" "artists")
+-- then: runSqlite "chinook.db" (selectToDataFrame [ArtistsName ==. Just "Accept"] [])
+@
+-}
+module DataFrame.IO.Persistent.Schema (
+ -- * Typed schema
+ declareTable,
+ declareTableWith,
+ declareTableFromSchema,
+
+ -- * Persistent entity generation
+ declareEntity,
+ declareEntityWith,
+
+ -- * Options + re-exports
+ DeclareOptions (..),
+ defaultDeclareOptions,
+ ColumnInfo (..),
+ HaskellType (..),
+) where
+
+import Data.List (partition)
+import Data.Text (Text)
+import qualified Data.Text as T
+import Language.Haskell.TH
+
+import Database.Persist.Quasi (lowerCaseSettings, parse)
+import Database.Persist.Quasi.Internal (UnboundEntityDef)
+import Database.Persist.TH (mkPersist, sqlSettings)
+
+import DataFrame.IO.Persistent.Schema.Common
+import DataFrame.IO.Persistent.Schema.Introspect
+import DataFrame.Typed.Types (Column)
+
+{- | Read @table@ from the SQLite database at @path@ and emit a typed schema
+synonym, @type \Schema = '[Column ...]@. Read into it with
+'DataFrame.IO.Persistent.Read.readTableTyped'.
+-}
+declareTable :: FilePath -> String -> Q [Dec]
+declareTable = declareTableWith defaultDeclareOptions
+
+-- | 'declareTable' with custom 'DeclareOptions'.
+declareTableWith :: DeclareOptions -> FilePath -> String -> Q [Dec]
+declareTableWith opts path table = do
+ cols <- introspectQ path table
+ warnOnNumeric cols
+ schemaDecl opts (T.pack table) cols
+
+{- | Like 'declareTable' but from an explicit column list, so __no database is
+opened at build time__ (use in CI / cross-compilation / sdist consumers).
+-}
+declareTableFromSchema :: [ColumnInfo] -> String -> Q [Dec]
+declareTableFromSchema cols name =
+ schemaDecl defaultDeclareOptions (T.pack name) cols
+
+-- | Emit @type \Schema = '[Column ...]@ for the selected columns.
+schemaDecl :: DeclareOptions -> Text -> [ColumnInfo] -> Q [Dec]
+schemaDecl opts table cols0 = do
+ let cols = selectColumns opts cols0
+ failOnEmpty table cols
+ pure [TySynD (mkName (schemaTypeNameOf opts table)) [] (schemaType opts cols)]
+
+failOnEmpty :: Text -> [ColumnInfo] -> Q ()
+failOnEmpty table [] =
+ fail ("declareTable: no columns selected for table " <> show (T.unpack table))
+failOnEmpty _ _ = pure ()
+
+schemaType :: DeclareOptions -> [ColumnInfo] -> Type
+schemaType opts cols =
+ buildSchemaType
+ [(ciName c, thType (resolvedType opts c), isNullable opts c) | c <- cols]
+
+-- | Fold @(name, type, nullable)@ triples into a promoted @'[Column name ty]@.
+buildSchemaType :: [(Text, Type, Bool)] -> Type
+buildSchemaType = foldr cons PromotedNilT
+ where
+ cons col acc = PromotedConsT `AppT` columnEntry col `AppT` acc
+
+columnEntry :: (Text, Type, Bool) -> Type
+columnEntry (name, ty, nullable) =
+ ConT ''Column
+ `AppT` LitT (StrTyLit (T.unpack name))
+ `AppT` wrapMaybe nullable ty
+
+wrapMaybe :: Bool -> Type -> Type
+wrapMaybe True ty = ConT ''Maybe `AppT` ty
+wrapMaybe False ty = ty
+
+{- | Read @table@ at compile time and generate a @persistent@ entity for it
+(via 'mkPersist'), giving the full @persistent@ typed-'Filter' DSL with no
+hand-written schema. Load it with 'DataFrame.IO.Persistent.Read.selectToDataFrame'.
+-}
+declareEntity :: FilePath -> String -> Q [Dec]
+declareEntity = declareEntityWith defaultDeclareOptions
+
+-- | 'declareEntity' with custom 'DeclareOptions'.
+declareEntityWith :: DeclareOptions -> FilePath -> String -> Q [Dec]
+declareEntityWith opts path table = do
+ cols <- introspectQ path table
+ let tableT = T.pack table
+ ent = T.pack (entityNameOf opts tableT)
+ dsl = renderEntityDsl opts ent tableT (selectColumns opts cols)
+ parseEntityDefs dsl >>= mkPersist sqlSettings
+
+parseEntityDefs :: Text -> Q [UnboundEntityDef]
+parseEntityDefs dsl = case snd (parse lowerCaseSettings [(Nothing, dsl)]) of
+ Left errs -> fail ("declareEntity: could not build entity definition:\n" <> show errs)
+ Right defs -> pure defs
+
+{- | Render the @persistLowerCase@ DSL text for an entity. An integer primary key
+becomes the @Id@ field; every other column is a field bound to its DB name.
+-}
+renderEntityDsl :: DeclareOptions -> Text -> Text -> [ColumnInfo] -> Text
+renderEntityDsl opts ent table cols =
+ T.unlines ((ent <> " sql=" <> table) : idLines <> map (renderField opts) fields)
+ where
+ (pks, others) = partition (isIntPk opts) cols
+ -- Rename persistent's default (ToBackendKey-enabled) key column rather than
+ -- declaring a natural key, so generated entities keep fromSqlKey/ToBackendKey.
+ (idLines, fields) = case pks of
+ [pk] -> ([" Id sql=" <> ciName pk], others)
+ _ -> ([], cols)
+
+isIntPk :: DeclareOptions -> ColumnInfo -> Bool
+isIntPk opts c = ciPk c && resolvedType opts c == HTInt
+
+renderField :: DeclareOptions -> ColumnInfo -> Text
+renderField opts c =
+ " "
+ <> fieldIdentifier (ciName c)
+ <> " "
+ <> persistType (resolvedType opts c)
+ <> " sql="
+ <> ciName c
+ <> (if isNullable opts c then " Maybe" else "")
diff --git a/src/DataFrame/IO/Persistent/Schema/Common.hs b/src/DataFrame/IO/Persistent/Schema/Common.hs
new file mode 100644
--- /dev/null
+++ b/src/DataFrame/IO/Persistent/Schema/Common.hs
@@ -0,0 +1,203 @@
+{-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE ScopedTypeVariables #-}
+{-# LANGUAGE TemplateHaskellQuotes #-}
+
+{- |
+Module : DataFrame.IO.Persistent.Schema.Common
+License : MIT
+
+Shared options, name mangling, type mapping, and compile-time introspection used
+by both the typed-schema splice (@declareTable@) and the entity splice
+(@declareEntity@) in "DataFrame.IO.Persistent.Schema".
+-}
+module DataFrame.IO.Persistent.Schema.Common (
+ -- * Options
+ DeclareOptions (..),
+ defaultDeclareOptions,
+
+ -- * Compile-time introspection
+ introspectQ,
+ selectColumns,
+
+ -- * Per-column resolution
+ resolvedType,
+ isNullable,
+
+ -- * Type mapping
+ thType,
+ persistType,
+
+ -- * Name helpers
+ pascalCase,
+ schemaTypeNameOf,
+ entityNameOf,
+ fieldIdentifier,
+
+ -- * Warnings
+ warnOnNumeric,
+) where
+
+import Control.Exception (SomeException, try)
+import Control.Monad (when)
+import Data.Char (isAlpha, isAlphaNum, toLower, toUpper)
+import Data.Either (fromRight)
+import Data.List (find, intercalate)
+import Data.Maybe (fromMaybe)
+import Data.Text (Text)
+import qualified Data.Text as T
+import Data.Time (Day, TimeOfDay, UTCTime)
+import Language.Haskell.TH (Q, Type (ConT), reportWarning)
+import Language.Haskell.TH.Syntax (addDependentFile, runIO)
+
+import Data.ByteString (ByteString)
+import DataFrame.IO.Persistent.Schema.Introspect
+
+{- | Knobs shared by 'DataFrame.IO.Persistent.Schema.declareTable' and
+'DataFrame.IO.Persistent.Schema.declareEntity'. Start from
+'defaultDeclareOptions' and override fields with record syntax.
+-}
+data DeclareOptions = DeclareOptions
+ { honorNotNull :: Bool
+ {- ^ @True@ (default): @NOT NULL@ columns are non-null, others @Maybe@.
+ @False@: every column is treated as nullable.
+ -}
+ , typeOverride :: Text -> Text -> Maybe HaskellType
+ {- ^ Given @(rawColumnName, declaredSqliteType)@, optionally override the
+ inferred Haskell type. Default: never override.
+ -}
+ , includeColumns :: Maybe [Text]
+ -- ^ Restrict to these raw columns (@Nothing@ = all).
+ , excludeColumns :: [Text]
+ -- ^ Drop these raw columns (applied after 'includeColumns').
+ , schemaTypeName :: Maybe String
+ -- ^ Override the generated schema type-synonym name (default @\Schema@).
+ , entityName :: Maybe String
+ -- ^ Override the generated entity name (default PascalCased table).
+ }
+
+-- | Sensible defaults: honor @NOT NULL@, no overrides, all columns, derived names.
+defaultDeclareOptions :: DeclareOptions
+defaultDeclareOptions =
+ DeclareOptions
+ { honorNotNull = True
+ , typeOverride = \_ _ -> Nothing
+ , includeColumns = Nothing
+ , excludeColumns = []
+ , schemaTypeName = Nothing
+ , entityName = Nothing
+ }
+
+{- | Introspect a table at compile time: register the DB as a dependency (so a
+schema change forces a rebuild), then fail helpfully if it can't be read.
+-}
+introspectQ :: FilePath -> String -> Q [ColumnInfo]
+introspectQ path table = do
+ addDependentFile path
+ outcome <- runIO (try (introspectTable path table))
+ case outcome of
+ Left (e :: SomeException) -> fail (openError path e)
+ Right [] -> failMissingTable path table
+ Right cols -> pure cols
+
+failMissingTable :: FilePath -> String -> Q a
+failMissingTable path table = do
+ names <-
+ runIO (try (introspectTableNames path) :: IO (Either SomeException [Text]))
+ fail (missingTableMessage path table (fromRight [] names))
+
+openError :: FilePath -> SomeException -> String
+openError path e =
+ "declareTable/declareEntity: could not read " <> path <> ": " <> show e
+
+missingTableMessage :: FilePath -> String -> [Text] -> String
+missingTableMessage path table names =
+ "declareTable/declareEntity: table "
+ <> show table
+ <> " not found in "
+ <> path
+ <> ".\n Available tables: "
+ <> intercalate ", " (map T.unpack names)
+ <> maybe
+ ""
+ (\s -> "\n (did you mean " <> show (T.unpack s) <> "?)")
+ (didYouMean table names)
+
+didYouMean :: String -> [Text] -> Maybe Text
+didYouMean target = find ((== map toLower target) . map toLower . T.unpack)
+
+-- | Apply 'includeColumns' / 'excludeColumns'.
+selectColumns :: DeclareOptions -> [ColumnInfo] -> [ColumnInfo]
+selectColumns opts = filter keep
+ where
+ keep c =
+ maybe True (ciName c `elem`) (includeColumns opts)
+ && ciName c `notElem` excludeColumns opts
+
+-- | The Haskell type for a column, honoring 'typeOverride'.
+resolvedType :: DeclareOptions -> ColumnInfo -> HaskellType
+resolvedType opts c =
+ fromMaybe
+ (inferType (ciDeclType c))
+ (typeOverride opts (ciName c) (ciDeclType c))
+
+-- | Whether a column should be modelled as nullable.
+isNullable :: DeclareOptions -> ColumnInfo -> Bool
+isNullable opts c = not (honorNotNull opts) || not (ciNotNull c)
+
+-- | The Template Haskell 'Type' for a 'HaskellType'.
+thType :: HaskellType -> Type
+thType HTInt = ConT ''Int
+thType HTDouble = ConT ''Double
+thType HTText = ConT ''Text
+thType HTBool = ConT ''Bool
+thType HTByteString = ConT ''ByteString
+thType HTDay = ConT ''Day
+thType HTUTCTime = ConT ''UTCTime
+thType HTTimeOfDay = ConT ''TimeOfDay
+
+-- | The @persistent@ DSL type name for a 'HaskellType'.
+persistType :: HaskellType -> Text
+persistType HTInt = "Int64"
+persistType HTDouble = "Double"
+persistType HTText = "Text"
+persistType HTBool = "Bool"
+persistType HTByteString = "ByteString"
+persistType HTDay = "Day"
+persistType HTUTCTime = "UTCTime"
+persistType HTTimeOfDay = "TimeOfDay"
+
+-- | Capitalize the first character (e.g. @"artists" -> "Artists"@).
+pascalCase :: Text -> String
+pascalCase t = case T.unpack t of
+ (c : cs) -> toUpper c : cs
+ [] -> []
+
+schemaTypeNameOf :: DeclareOptions -> Text -> String
+schemaTypeNameOf opts table = fromMaybe (pascalCase table <> "Schema") (schemaTypeName opts)
+
+entityNameOf :: DeclareOptions -> Text -> String
+entityNameOf opts table = fromMaybe (pascalCase table) (entityName opts)
+
+-- | Turn a raw column name into a valid lowercase-initial @persistent@ field name.
+fieldIdentifier :: Text -> Text
+fieldIdentifier raw = T.pack (lowerFirst (ensureLetter cleaned))
+ where
+ cleaned = filter (\ch -> isAlphaNum ch || ch == '_') (T.unpack raw)
+ lowerFirst (c : cs) = toLower c : cs
+ lowerFirst [] = "field"
+ ensureLetter s@(c : _) | isAlpha c = s
+ ensureLetter s = 'f' : s
+
+-- | Warn (at splice time) about @NUMERIC@/@DECIMAL@ columns mapped to 'Double'.
+warnOnNumeric :: [ColumnInfo] -> Q ()
+warnOnNumeric = mapM_ warn1
+ where
+ warn1 c = when (isNumeric (ciDeclType c)) (reportWarning (numericMessage c))
+ isNumeric t = any (`T.isInfixOf` T.toUpper t) ["NUMERIC", "DECIMAL"]
+ numericMessage c =
+ "declareTable: column "
+ <> show (T.unpack (ciName c))
+ <> " has SQLite type "
+ <> show (T.unpack (ciDeclType c))
+ <> "; mapping to Double may lose precision "
+ <> "(override with typeOverride)."
diff --git a/src/DataFrame/IO/Persistent/Schema/Introspect.hs b/src/DataFrame/IO/Persistent/Schema/Introspect.hs
new file mode 100644
--- /dev/null
+++ b/src/DataFrame/IO/Persistent/Schema/Introspect.hs
@@ -0,0 +1,211 @@
+{-# LANGUAGE FlexibleContexts #-}
+{-# LANGUAGE OverloadedStrings #-}
+
+{- |
+Module : DataFrame.IO.Persistent.Schema.Introspect
+License : MIT
+
+Compile-time-safe SQLite schema introspection. Everything here is plain 'IO'
+(callable from a Template Haskell @runIO@ in "DataFrame.IO.Persistent.Schema")
+plus a pure SQLite-affinity → 'HaskellType' mapping. No Template Haskell and no
+DataFrame dependencies live here, so the logic can be unit-tested directly.
+-}
+module DataFrame.IO.Persistent.Schema.Introspect (
+ -- * Column metadata
+ ColumnInfo (..),
+ HaskellType (..),
+
+ -- * Introspection (IO)
+ introspectTable,
+ introspectTableConn,
+ introspectTableNames,
+ introspectTableNamesConn,
+
+ -- * Type inference
+ inferType,
+ haskellTypeName,
+
+ -- * Raw-query helper (shared with the runtime reader)
+ runRawRows,
+ quoteIdent,
+) where
+
+import Control.Monad.Trans.Reader (ReaderT)
+import Control.Monad.Trans.Resource (MonadResource)
+import Data.Conduit (runConduit, (.|))
+import qualified Data.Conduit.List as CL
+import Data.List (sortOn)
+import Data.Text (Text)
+import qualified Data.Text as T
+import Database.Persist (PersistValue (..))
+import Database.Persist.Sql (SqlBackend, rawQuery)
+import Database.Persist.SqlBackend (getRDBMS)
+import Database.Persist.Sqlite (runSqlite)
+
+-- | A single column as reported by @PRAGMA table_info@.
+data ColumnInfo = ColumnInfo
+ { ciName :: Text
+ -- ^ Raw column name as stored in SQLite (e.g. @"ArtistId"@).
+ , ciDeclType :: Text
+ -- ^ Declared type string (e.g. @"NVARCHAR(120)"@); may be empty.
+ , ciNotNull :: Bool
+ -- ^ Whether the column carries a @NOT NULL@ constraint.
+ , ciPk :: Bool
+ -- ^ Whether the column participates in the primary key.
+ }
+ deriving (Eq, Show)
+
+{- | The closed set of Haskell element types we map SQLite columns onto. Every
+constructor corresponds to a type that is both 'Database.Persist.PersistField'
+and 'DataFrame.Internal.Column.Columnable'.
+-}
+data HaskellType
+ = HTInt
+ | HTDouble
+ | HTText
+ | HTBool
+ | HTByteString
+ | HTDay
+ | HTUTCTime
+ | HTTimeOfDay
+ deriving (Eq, Show)
+
+{- | Map a declared SQLite type to a 'HaskellType' using SQLite's own
+(case-insensitive, substring) type-affinity rules, with a few semantic
+refinements for dates/booleans. Unknown or empty declarations fall back the
+same way SQLite does (BLOB affinity for empty, numeric otherwise).
+-}
+inferType :: Text -> HaskellType
+inferType declared
+ | "INT" `inUpper` declared = HTInt
+ | anyIn ["CHAR", "CLOB", "TEXT"] declared = HTText
+ | anyIn ["BLOB", "BYTEA"] declared || T.null (T.strip declared) = HTByteString
+ | anyIn ["REAL", "FLOA", "DOUB"] declared = HTDouble
+ | "BOOL" `inUpper` declared = HTBool
+ | anyIn ["DATETIME", "TIMESTAMP"] declared = HTUTCTime
+ | "DATE" `inUpper` declared = HTDay
+ | "TIME" `inUpper` declared = HTTimeOfDay
+ | otherwise = HTDouble
+
+-- | A display name for a 'HaskellType' (the element type, without nullability).
+haskellTypeName :: HaskellType -> Text
+haskellTypeName HTInt = "Int"
+haskellTypeName HTDouble = "Double"
+haskellTypeName HTText = "Text"
+haskellTypeName HTBool = "Bool"
+haskellTypeName HTByteString = "ByteString"
+haskellTypeName HTDay = "Day"
+haskellTypeName HTUTCTime = "UTCTime"
+haskellTypeName HTTimeOfDay = "TimeOfDay"
+
+inUpper :: Text -> Text -> Bool
+inUpper needle hay = needle `T.isInfixOf` T.toUpper hay
+
+anyIn :: [Text] -> Text -> Bool
+anyIn needles hay = any (`inUpper` hay) needles
+
+{- | Read the columns of @table@ from the SQLite database at @path@, ordered by
+their @PRAGMA table_info@ @cid@ (i.e. declaration order).
+-}
+introspectTable :: FilePath -> String -> IO [ColumnInfo]
+introspectTable path table =
+ runSqlite (T.pack path) (introspectTableConn (T.pack table))
+
+{- | 'introspectTable' over an existing connection. Backend-agnostic: SQLite uses
+@PRAGMA table_info@; every other backend (PostgreSQL, MySQL, etc.) uses the
+standard @information_schema.columns@.
+-}
+introspectTableConn ::
+ (MonadResource m) => Text -> ReaderT SqlBackend m [ColumnInfo]
+introspectTableConn table = do
+ rdbms <- getRDBMS
+ if rdbms == "sqlite" then sqlitePragmaColumns table else infoSchemaColumns table
+
+sqlitePragmaColumns ::
+ (MonadResource m) => Text -> ReaderT SqlBackend m [ColumnInfo]
+sqlitePragmaColumns table = do
+ rows <- runRawRows ("PRAGMA table_info(" <> quoteIdent table <> ")") []
+ pure (map snd (sortOn fst (map parsePragmaRow rows)))
+
+infoSchemaColumns ::
+ (MonadResource m) => Text -> ReaderT SqlBackend m [ColumnInfo]
+infoSchemaColumns table = do
+ rows <- runRawRows infoSchemaColumnQuery [PersistText table]
+ pks <- runRawRows infoSchemaPkQuery [PersistText table]
+ let pkNames = [pvText n | (n : _) <- pks]
+ pure (map (parseInfoSchemaRow pkNames) rows)
+
+infoSchemaColumnQuery :: Text
+infoSchemaColumnQuery =
+ "SELECT column_name, data_type, is_nullable FROM information_schema.columns \
+ \WHERE table_name = ? AND table_schema NOT IN ('pg_catalog', 'information_schema') \
+ \ORDER BY ordinal_position"
+
+infoSchemaPkQuery :: Text
+infoSchemaPkQuery =
+ "SELECT kcu.column_name FROM information_schema.table_constraints tc \
+ \JOIN information_schema.key_column_usage kcu ON kcu.constraint_name = tc.constraint_name \
+ \WHERE tc.table_name = ? AND tc.constraint_type = 'PRIMARY KEY'"
+
+-- | List the user tables (excluding SQLite internal tables) in @path@.
+introspectTableNames :: FilePath -> IO [Text]
+introspectTableNames path = runSqlite (T.pack path) introspectTableNamesConn
+
+-- | 'introspectTableNames' over an existing connection (backend-agnostic).
+introspectTableNamesConn :: (MonadResource m) => ReaderT SqlBackend m [Text]
+introspectTableNamesConn = do
+ rdbms <- getRDBMS
+ rows <-
+ runRawRows
+ (if rdbms == "sqlite" then sqliteTableQuery else infoSchemaTableQuery)
+ []
+ pure [pvText n | (n : _) <- rows]
+
+sqliteTableQuery :: Text
+sqliteTableQuery =
+ "SELECT name FROM sqlite_master WHERE type = 'table' \
+ \AND name NOT LIKE 'sqlite_%' ORDER BY name"
+
+infoSchemaTableQuery :: Text
+infoSchemaTableQuery =
+ "SELECT table_name FROM information_schema.tables \
+ \WHERE table_schema NOT IN ('pg_catalog', 'information_schema') \
+ \AND table_type = 'BASE TABLE' ORDER BY table_name"
+
+-- | Run a raw SQL query and collect every row's column values.
+runRawRows ::
+ (MonadResource m) =>
+ Text -> [PersistValue] -> ReaderT SqlBackend m [[PersistValue]]
+runRawRows sql params = runConduit (rawQuery sql params .| CL.consume)
+
+parsePragmaRow :: [PersistValue] -> (Int, ColumnInfo)
+parsePragmaRow [cid, name, ctype, notnull, _dflt, pk] =
+ ( pvInt cid
+ , ColumnInfo (pvText name) (pvText ctype) (pvInt notnull == 1) (pvInt pk > 0)
+ )
+parsePragmaRow other =
+ error ("introspectTable: unexpected PRAGMA table_info row: " <> show other)
+
+{- | Parse an @information_schema.columns@ row @(name, data_type, is_nullable)@,
+marking 'ciPk' from the table's primary-key column names.
+-}
+parseInfoSchemaRow :: [Text] -> [PersistValue] -> ColumnInfo
+parseInfoSchemaRow pkNames [name, dtype, nullable] =
+ ColumnInfo nm (pvText dtype) (pvText nullable == "NO") (nm `elem` pkNames)
+ where
+ nm = pvText name
+parseInfoSchemaRow _ other =
+ error ("introspectTable: unexpected information_schema row: " <> show other)
+
+pvText :: PersistValue -> Text
+pvText (PersistText t) = t
+pvText PersistNull = ""
+pvText other = T.pack (show other)
+
+pvInt :: PersistValue -> Int
+pvInt (PersistInt64 i) = fromIntegral i
+pvInt _ = 0
+
+-- | Double-quote a SQLite identifier, escaping embedded quotes.
+quoteIdent :: Text -> Text
+quoteIdent t = "\"" <> T.replace "\"" "\"\"" t <> "\""
diff --git a/src/DataFrame/IO/Persistent/TH.hs b/src/DataFrame/IO/Persistent/TH.hs
--- a/src/DataFrame/IO/Persistent/TH.hs
+++ b/src/DataFrame/IO/Persistent/TH.hs
@@ -14,20 +14,16 @@
derivePersistentDataFrame,
) where
-import Control.Monad (forM, when)
-import Data.List (foldl')
-import Data.Text (Text)
+import Control.Monad (forM)
+import Data.Char
import qualified Data.Text as T
import qualified Data.Vector as V
import DataFrame.Functions (col)
import DataFrame.IO.Persistent
-import qualified DataFrame.Internal.Column as DFCol
import DataFrame.Internal.Expression
import Database.Persist
-import Database.Persist.Sql (fromSqlKey)
-import Database.Persist.TH
import Language.Haskell.TH
-import Language.Haskell.TH.Syntax (Lift, lift)
+import Language.Haskell.TH.Syntax (lift)
import Debug.Trace (trace)
@@ -71,13 +67,13 @@
dataframeExprs <- forM fields $ \(raw, _, ty) -> do
let nm = camelToSnake (nameBase raw)
let colName = camelToSnake (drop (length entityNameStr) (nameBase raw))
- trace ((nm <> " :: Expr " <> (show ty))) pure ()
+ trace (nm <> " :: Expr " <> show ty) pure ()
let n = mkName nm
sig <- sigD n [t|Expr $(pure ty)|]
val <- valD (varP n) (normalB [|col $(lift colName)|]) []
pure [sig, val]
- return ([instanceDec] ++ concat dataframeExprs)
+ return (instanceDec : concat dataframeExprs)
_ ->
fail $
"deriveEntityToDataFrame: " ++ show entityName ++ " must be a record type"
@@ -190,5 +186,5 @@
go (c : cs)
| isUpper c = '_' : toLower c : go cs
| otherwise = c : go cs
- isUpper c = c >= 'A' && c <= 'Z'
+ isUpper = isAsciiUpper
toLower c = if isUpper c then toEnum (fromEnum c + 32) else c
diff --git a/tests/Main.hs b/tests/Main.hs
--- a/tests/Main.hs
+++ b/tests/Main.hs
@@ -3,12 +3,13 @@
import qualified System.Exit as Exit
import PersistentTests
+import SchemaTHTests
import Test.HUnit
main :: IO ()
main = do
- result <- runTestTT persistentTests
+ result <- runTestTT (TestList [persistentTests, schemaThTests])
if failures result > 0 || errors result > 0
then Exit.exitFailure
else Exit.exitSuccess
diff --git a/tests/PersistentTests.hs b/tests/PersistentTests.hs
--- a/tests/PersistentTests.hs
+++ b/tests/PersistentTests.hs
@@ -13,6 +13,7 @@
{-# LANGUAGE TemplateHaskell #-}
{-# LANGUAGE TypeApplications #-}
{-# LANGUAGE TypeFamilies #-}
+{-# LANGUAGE TypeOperators #-}
{-# LANGUAGE UndecidableInstances #-}
module PersistentTests where
@@ -36,7 +37,7 @@
import Test.HUnit
import UnliftIO.Resource (ResourceT)
-import DataFrame.Functions ((.<))
+import DataFrame.Functions ((.<.))
-- Define test entities
share
@@ -109,7 +110,7 @@
assertBool "Has active column" ("active" `elem` cols)
-- Check data values
- let names = V.toList (DF.columnAsVector test_user_name df)
+ let names = DF.columnAsList test_user_name df
assertEqual "Names match" ["Alice", "Bob", "Charlie", "Diana"] names
@@ -125,7 +126,7 @@
print cols
assertBool "Has id column" ("id" `elem` cols)
assertBool "Has name column" ("name" `elem` cols)
- let names = V.toList (DF.columnAsVector artist_name (DF.take 5 df))
+ let names = DF.columnAsList artist_name (DF.take 5 df)
assertEqual
"Names match"
@@ -144,7 +145,7 @@
assertEqual "Active users count" 3 (nRows df)
-- Check all loaded users are active
- let activeFlags = V.toList $ DF.columnAsVector test_user_active df
+ let activeFlags = DF.columnAsList test_user_active df
assertBool "All active" (and activeFlags)
-- Test custom configuration
@@ -196,15 +197,15 @@
assertEqual "Young users count" 2 (nRows youngUsers)
-- Sort operation
- let sorted = DF.sortBy [DF.Asc "age"] df
- let ages = V.toList $ DF.columnAsVector test_user_age sorted
+ let sorted = DF.sortBy [DF.Asc (F.col @Int "age")] df
+ let ages = DF.columnAsList test_user_age sorted
assertEqual "Ages sorted" [25, 28, 30, 35] ages
-- Derive column
let withAgeGroup =
DF.derive @Text
"age_group"
- (F.ifThenElse (test_user_age .< 30) "young" "adult")
+ (F.ifThenElse (test_user_age .<. 30) "young" "adult")
df
assertEqual "Has age_group column" 5 (length $ DF.columnNames withAgeGroup)
@@ -239,7 +240,7 @@
assertEqual "Empty DataFrame" 0 (nRows df)
-- Should still have column names
let cols = DF.columnNames df
- assertBool "Has columns" (length cols > 0)
+ assertBool "Has columns" (not (null cols))
-- Test persistFieldsToColumns function
testPersistFieldsToColumns :: Test
diff --git a/tests/SchemaTHTests.hs b/tests/SchemaTHTests.hs
new file mode 100644
--- /dev/null
+++ b/tests/SchemaTHTests.hs
@@ -0,0 +1,128 @@
+{-# LANGUAGE DataKinds #-}
+{-# LANGUAGE DerivingStrategies #-}
+{-# LANGUAGE FlexibleContexts #-}
+{-# LANGUAGE FlexibleInstances #-}
+{-# LANGUAGE GADTs #-}
+{-# LANGUAGE GeneralizedNewtypeDeriving #-}
+{-# LANGUAGE MultiParamTypeClasses #-}
+{-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE QuasiQuotes #-}
+{-# LANGUAGE StandaloneDeriving #-}
+{-# LANGUAGE TemplateHaskell #-}
+{-# LANGUAGE TypeApplications #-}
+{-# LANGUAGE TypeFamilies #-}
+{-# LANGUAGE TypeOperators #-}
+{-# LANGUAGE UndecidableInstances #-}
+
+{- |
+End-to-end tests for the compile-time SQLite schema splices and the runtime
+readers, exercised against the committed @data/chinook.db@ fixture.
+-}
+module SchemaTHTests (schemaThTests) where
+
+import Control.Monad.IO.Class (liftIO)
+import Data.Function ((&))
+import Data.Int (Int64)
+import Data.Maybe (listToMaybe)
+import Data.Text (Text)
+import qualified Data.Text as T
+import Database.Persist (Filter, (==.))
+import Database.Persist.Sqlite (runSqlite)
+import Test.HUnit
+
+import qualified DataFrame as DF
+import DataFrame.IO.Persistent (nRows)
+import DataFrame.IO.Persistent.Read
+import DataFrame.IO.Persistent.Schema
+import qualified DataFrame.Typed as DT
+
+-- | Typed reader + schema for the @artists@ table (read at compile time).
+$(declareTable "./data/chinook.db" "artists")
+
+-- | A @persistent@ entity for the @albums@ table (generated from the live DB).
+$(declareEntity "./data/chinook.db" "albums")
+
+chinook :: T.Text
+chinook = "./data/chinook.db"
+
+-- Tier 0: runtime readers --------------------------------------------------
+
+testReadTable :: Test
+testReadTable = TestCase $ do
+ df <- readTable "./data/chinook.db" "artists"
+ assertEqual "row count" 275 (nRows df)
+ let cols = DF.columnNames df
+ assertBool "has raw ArtistId" ("ArtistId" `elem` cols)
+ assertBool "has raw Name" ("Name" `elem` cols)
+
+testReadTableWith :: Test
+testReadTableWith = TestCase $ do
+ df <- readTableWith "./data/chinook.db" "artists" (allRows & limit 5)
+ assertEqual "limit pushed down" 5 (nRows df)
+
+testReadSql :: Test
+testReadSql = TestCase $ do
+ df <- readSql "./data/chinook.db" "SELECT Name FROM artists LIMIT 5"
+ assertEqual "raw query rows" 5 (nRows df)
+
+testListTables :: Test
+testListTables = TestCase $ do
+ ts <- listTables "./data/chinook.db"
+ assertBool "has artists" ("artists" `elem` ts)
+ assertBool "has albums" ("albums" `elem` ts)
+
+testDescribeTable :: Test
+testDescribeTable = TestCase $ do
+ d <- describeTable "./data/chinook.db" "artists"
+ assertEqual
+ "describe-style columns"
+ ["Column Name", "Type", "SQLite Type", "Nullable", "Primary Key"]
+ (DF.columnNames d)
+ assertEqual "one row per column" 2 (nRows d)
+
+-- Tier 1: typed splice -----------------------------------------------------
+
+testDeclareTableTyped :: Test
+testDeclareTableTyped = TestCase $ do
+ -- declareTable emits only the ArtistsSchema type; read into it generically:
+ tdf <- readTableTyped @ArtistsSchema "./data/chinook.db" "artists"
+ assertEqual "typed row count" 275 (nRows (DT.thaw tdf))
+ -- @col \@"Name"@ here is checked against the generated ArtistsSchema:
+ let names = DT.columnAsList @"Name" tdf
+ assertEqual "first artist" (Just (Just "AC/DC")) (listToMaybe names)
+
+-- Tier 2: persistent entity generation -------------------------------------
+
+testDeclareEntity :: Test
+testDeclareEntity = TestCase $ runSqlite chinook $ do
+ df <- selectToDataFrame ([] :: [Filter Albums]) []
+ liftIO $ assertEqual "all albums" 347 (nRows df)
+
+testDeclareEntityFilter :: Test
+testDeclareEntityFilter = TestCase $ runSqlite chinook $ do
+ df <- selectToDataFrame [AlbumsArtistId ==. 1] []
+ liftIO $ assertEqual "AC/DC albums" 2 (nRows df)
+
+-- Hand-off: raw read and entity read share one connection ------------------
+
+testHandOff :: Test
+testHandOff = TestCase $ runSqlite chinook $ do
+ artists <- readTableConn "artists"
+ albums <- selectToDataFrame ([] :: [Filter Albums]) []
+ liftIO $ do
+ assertEqual "artists via raw conn" 275 (nRows artists)
+ assertEqual "albums via entity" 347 (nRows albums)
+
+schemaThTests :: Test
+schemaThTests =
+ TestList
+ [ TestLabel "readTable" testReadTable
+ , TestLabel "readTableWith limit" testReadTableWith
+ , TestLabel "readSql" testReadSql
+ , TestLabel "listTables" testListTables
+ , TestLabel "describeTable" testDescribeTable
+ , TestLabel "declareTable (typed)" testDeclareTableTyped
+ , TestLabel "declareEntity" testDeclareEntity
+ , TestLabel "declareEntity filter" testDeclareEntityFilter
+ , TestLabel "dataframe/persistent hand-off" testHandOff
+ ]