]> git.immae.eu Git - github/fretlink/tap-google-sheets.git/commitdiff
Initial commit
authorJeff Huth <jeff.huth@bytecode.io>
Thu, 14 Nov 2019 01:03:56 +0000 (17:03 -0800)
committerJeff Huth <jeff.huth@bytecode.io>
Thu, 14 Nov 2019 01:03:56 +0000 (17:03 -0800)
Discovery mode works. Still working on normal sync.

18 files changed:
.gitignore [new file with mode: 0644]
CHANGELOG.md [new file with mode: 0644]
LICENSE [new file with mode: 0644]
MANIFEST.in [new file with mode: 0644]
README.md
config.json.example [new file with mode: 0644]
setup.py [new file with mode: 0644]
state.json.example [new file with mode: 0644]
tap_google_sheets/__init__.py [new file with mode: 0644]
tap_google_sheets/client.py [new file with mode: 0644]
tap_google_sheets/discover.py [new file with mode: 0644]
tap_google_sheets/schema.py [new file with mode: 0644]
tap_google_sheets/schemas/file_metadata.json [new file with mode: 0644]
tap_google_sheets/schemas/sheet_metadata.json [new file with mode: 0644]
tap_google_sheets/schemas/sheets_loaded.json [new file with mode: 0644]
tap_google_sheets/schemas/spreadsheet_metadata.json [new file with mode: 0644]
tap_google_sheets/streams.py [new file with mode: 0644]
tap_google_sheets/sync.py [new file with mode: 0644]

diff --git a/.gitignore b/.gitignore
new file mode 100644 (file)
index 0000000..d9633da
--- /dev/null
@@ -0,0 +1,22 @@
+/.vscode/*
+/.ipynb_checkpoints/*
+config/
+virtualenvs/
+*catalog*.json
+*config*.json
+*state*.json
+target*.json
+*.sublime-*
+.python-version
+singer-check-tap-data
+*.pyc
+*.egg-info
+dist/
+__pycache__/
+venv/
+build/
+tap_google_sheets/.vscode/settings.json
+*.ipynb
+.DS_Store
+test_configuration.py
+tap_target_commands.sh
diff --git a/CHANGELOG.md b/CHANGELOG.md
new file mode 100644 (file)
index 0000000..d58f396
--- /dev/null
@@ -0,0 +1,4 @@
+# Changelog
+
+## 0.0.1
+  * Initial commit
diff --git a/LICENSE b/LICENSE
new file mode 100644 (file)
index 0000000..0ad25db
--- /dev/null
+++ b/LICENSE
@@ -0,0 +1,661 @@
+                    GNU AFFERO GENERAL PUBLIC LICENSE
+                       Version 3, 19 November 2007
+
+ Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+                            Preamble
+
+  The GNU Affero General Public License is a free, copyleft license for
+software and other kinds of works, specifically designed to ensure
+cooperation with the community in the case of network server software.
+
+  The licenses for most software and other practical works are designed
+to take away your freedom to share and change the works.  By contrast,
+our General Public Licenses are 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.
+
+  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.
+
+  Developers that use our General Public Licenses protect your rights
+with two steps: (1) assert copyright on the software, and (2) offer
+you this License which gives you legal permission to copy, distribute
+and/or modify the software.
+
+  A secondary benefit of defending all users' freedom is that
+improvements made in alternate versions of the program, if they
+receive widespread use, become available for other developers to
+incorporate.  Many developers of free software are heartened and
+encouraged by the resulting cooperation.  However, in the case of
+software used on network servers, this result may fail to come about.
+The GNU General Public License permits making a modified version and
+letting the public access it on a server without ever releasing its
+source code to the public.
+
+  The GNU Affero General Public License is designed specifically to
+ensure that, in such cases, the modified source code becomes available
+to the community.  It requires the operator of a network server to
+provide the source code of the modified version running there to the
+users of that server.  Therefore, public use of a modified version, on
+a publicly accessible server, gives the public access to the source
+code of the modified version.
+
+  An older license, called the Affero General Public License and
+published by Affero, was designed to accomplish similar goals.  This is
+a different license, not a version of the Affero GPL, but Affero has
+released a new version of the Affero GPL which permits relicensing under
+this license.
+
+  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 Affero 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. Remote Network Interaction; Use with the GNU General Public License.
+
+  Notwithstanding any other provision of this License, if you modify the
+Program, your modified version must prominently offer all users
+interacting with it remotely through a computer network (if your version
+supports such interaction) an opportunity to receive the Corresponding
+Source of your version by providing access to the Corresponding Source
+from a network server at no charge, through some standard or customary
+means of facilitating copying of software.  This Corresponding Source
+shall include the Corresponding Source for any work covered by version 3
+of the GNU General Public License that is incorporated pursuant to the
+following paragraph.
+
+  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 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 work with which it is combined will remain governed by version
+3 of the GNU General Public License.
+
+  14. Revised Versions of this License.
+
+  The Free Software Foundation may publish revised and/or new versions of
+the GNU Affero 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 Affero 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 Affero 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 Affero 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.
+
+    <one line to give the program's name and a brief idea of what it does.>
+    Copyright (C) <year>  <name of author>
+
+    This program is free software: you can redistribute it and/or modify
+    it under the terms of the GNU Affero 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 Affero General Public License for more details.
+
+    You should have received a copy of the GNU Affero General Public License
+    along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+Also add information on how to contact you by electronic and paper mail.
+
+  If your software can interact with users remotely through a computer
+network, you should also make sure that it provides a way for users to
+get its source.  For example, if your program is a web application, its
+interface could display a "Source" link that leads users to an archive
+of the code.  There are many ways you could offer source, and different
+solutions will be better for different programs; see section 13 for the
+specific requirements.
+
+  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 AGPL, see
+<https://www.gnu.org/licenses/>.
diff --git a/MANIFEST.in b/MANIFEST.in
new file mode 100644 (file)
index 0000000..32a5d00
--- /dev/null
@@ -0,0 +1,2 @@
+include LICENSE
+include tap_google_sheets/schemas/*.json
index db00e68d3f4b020e863602bdbe674776623df287..6fcb0b95d86e7743fc1ce251825ac5dcc4082ba3 100644 (file)
--- a/README.md
+++ b/README.md
@@ -1 +1,172 @@
-tap-google-sheets
+# tap-google-sheets
+
+This is a [Singer](https://singer.io) tap that produces JSON-formatted data
+following the [Singer
+spec](https://github.com/singer-io/getting-started/blob/master/SPEC.md).
+
+This tap:
+
+- Pulls raw data from the [Google Sheets v4 API](https://developers.google.com/sheets/api)
+- Extracts the following endpoints:
+  - [Spreadsheet File](https://developers.google.com/drive/api/v3/reference/files/get)
+  - [Spreadsheet Metadata, Sheets, Headers](https://developers.google.com/sheets/api/reference/rest/v4/spreadsheets/get)
+  - [Spreadsheet Values](https://developers.google.com/sheets/api/reference/rest/v4/spreadsheets.values/get)
+- For each Sheet:
+  - Outputs the schema for each resource (based on the column header and datatypes of first row of data)
+  - Outputs a record for all columns with column headers, and for each row of data until it reaches an empty row
+
+## API Endpoints
+[**file (GET)**](https://developers.google.com/drive/api/v3/reference/files/get)
+- Endpoint: https://www.googleapis.com/drive/v3/files/${spreadsheet_id}?fields=id,name,createdTime,modifiedTime,version
+- Primary keys: id
+- Replication strategy: Full (GET file audit data for spreadsheet_id in config)
+- Process/Transformations: Replicate Data if Modified
+
+[**metadata (GET)**](https://developers.google.com/drive/api/v3/reference/files/get)
+- Endpoint: https://sheets.googleapis.com/v4/spreadsheets/${spreadsheet_id}?includeGridData=true&ranges=1:2
+- This endpoint eturns spreadsheet metadata, sheet metadata, and value metadata (data type information)
+- Primary keys: spreadsheetId, title, field_name
+- Foreign keys: None
+- Replication strategy: Full (get and replace file metadata for spreadshee_id in config)
+- Process/Transformations:
+  - Verify Sheets: Check sheets exist (compared to catalog) and check gridProperties (available area)
+    - sheetId, title, index, gridProperties (rowCount, columnCount)
+  - Verify Field Headers (1st row): Check field headers exist (compared to catalog), missing headers (columns to skip), column order/position, and column uniqueness
+    - Header's field_name, position: data.rowData[0].values[i].formattedValue
+  - Create/Verify Datatypes (2nd row): 
+    - Row 2's datatype, format: data.rowData[1].values[i]
+      - First check:
+        - [effectiveValue: key](https://developers.google.com/sheets/api/reference/rest/v4/spreadsheets/other#ExtendedValue)
+          - Valid types: numberValue, stringValue, boolValue
+          - Invalid types: formulaValue, errorValue
+      - Then check:
+        - [effectiveFormat.numberFormat.type](https://developers.google.com/sheets/api/reference/rest/v4/spreadsheets/cells#NumberFormatType)
+          - Valid types: UNEPECIFIED, TEXT, NUMBER, PERCENT, CURRENCY, DATE, TIME, DATE_TIME, SCIENTIFIC
+          - If DATE or DATE_TIME, set JSON schema datatype = string and format = date-time
+        - [effectiveFormat.numberFormat.pattern](https://developers.google.com/sheets/api/reference/rest/v4/spreadsheets/cells#NumberFormat)
+
+[**values (GET)**](https://developers.google.com/sheets/api/reference/rest/v4/spreadsheets.values/get)
+- Endpoint: https://sheets.googleapis.com/v4/spreadsheets/${spreadsheet_id}/values/'${sheet_name}'!${row_range}?dateTimeRenderOption=SERIAL_NUMBER&valueRenderOption=UNFORMATTED_VALUE&majorDimension=ROWS
+- This endpoint loops through sheets and row ranges to get the [unformatted values](https://developers.google.com/sheets/api/reference/rest/v4/ValueRenderOption) (effective values only), dates and datetimes as [serial numbers](https://developers.google.com/sheets/api/reference/rest/v4/DateTimeRenderOption)
+- Primary keys: row
+- Replication strategy: Full (GET file audit data for spreadsheet_id in config)
+- Process/Transformations:
+  - Loop through sheets (compared to catalog selection)
+    - Send metadata for sheet
+  - Loop through ranges of rows until reaching empty row or area max row (from sheet metadata)
+  - Transform values, if necessary (dates, date-times, boolean, integer, numers)
+  - Process/send records
+
+## Authentication
+The [**Google Sheets Setup & Authentication**](https://drive.google.com/open?id=1FojlvtLwS0-BzGS37R0jEXtwSHqSiO1Uw-7RKQQO-C4) Google Doc provides instructions show how to configure the Google Cloud API credentials to enable Google Drive and Google Sheets APIs, configure Google Cloud to authorize/verify your domain ownership, generate an API key (client_id, client_secret), authenticate and generate a refresh_token, and prepare your tap config.json with the necessary parameters.
+- Enable Googe Drive APIs and Authorization Scope: https://www.googleapis.com/auth/drive.metadata.readonly
+- Enable Google Sheets API and Authorization Scope: https://www.googleapis.com/auth/spreadsheets.readonly
+- Tap config.json parameters:
+  - client_id: identifies your application
+  - client_secret: authenticates your application
+  - refresh_token: generates an access token to authorize your session
+  - spreadsheet_id: unique identifier for each spreadsheet in Google Drive
+  - start_date: absolute minimum start date to check file modified
+  - user_agent: tap-name and email address; identifies your application in the Remote API server logs
+
+## Quick Start
+
+1. Install
+
+    Clone this repository, and then install using setup.py. We recommend using a virtualenv:
+
+    ```bash
+    > virtualenv -p python3 venv
+    > source venv/bin/activate
+    > python setup.py install
+    OR
+    > cd .../tap-google-sheets
+    > pip install .
+    ```
+2. Dependent libraries
+    The following dependent libraries were installed.
+    ```bash
+    > pip install target-json
+    > pip install target-stitch
+    > pip install singer-tools
+    > pip install singer-python
+    ```
+    - [singer-tools](https://github.com/singer-io/singer-tools)
+    - [target-stitch](https://github.com/singer-io/target-stitch)
+
+3. Create your tap's `config.json` file. Include the client_id, client_secret, refresh_token, site_urls (website URL properties in a comma delimited list; do not include the domain-level property in the list), start_date (UTC format), and user_agent (tap name with the api user email address).
+
+    ```json
+    {
+        "client_id": "YOUR_CLIENT_ID",
+        "client_secret": "YOUR_CLIENT_SECRET",
+        "refresh_token": "YOUR_REFRESH_TOKEN",
+        "spreadsheet_id": "YOUR_GOOGLE_SPREADSHEET_ID",
+        "start_date": "2019-01-01T00:00:00Z",
+        "user_agent": "tap-google-sheets <api_user_email@example.com>"
+    }
+    ```
+    
+    Optionally, also create a `state.json` file. `currently_syncing` is an optional attribute used for identifying the last object to be synced in case the job is interrupted mid-stream. The next run would begin where the last job left off.
+    Only the `performance_reports` uses a bookmark. The date-time bookmark is stored in a nested structure based on the endpoint, site, and sub_type.
+
+    ```json
+    {
+        "currently_syncing": "file_metadata",
+        "bookmarks": {
+            "file_metadata": "2019-09-27T22:34:39.000000Z",
+            "sheets_loaded": "2019-09-27T21:55:42.000000Z"
+        }
+    }
+    ```
+
+4. Run the Tap in Discovery Mode
+    This creates a catalog.json for selecting objects/fields to integrate:
+    ```bash
+    tap-google-sheets --config config.json --discover > catalog.json
+    ```
+   See the Singer docs on discovery mode
+   [here](https://github.com/singer-io/getting-started/blob/master/docs/DISCOVERY_MODE.md#discovery-mode).
+
+5. Run the Tap in Sync Mode (with catalog) and [write out to state file](https://github.com/singer-io/getting-started/blob/master/docs/RUNNING_AND_DEVELOPING.md#running-a-singer-tap-with-a-singer-target)
+
+    For Sync mode:
+    ```bash
+    > tap-google-sheets --config tap_config.json --catalog catalog.json > state.json
+    > tail -1 state.json > state.json.tmp && mv state.json.tmp state.json
+    ```
+    To load to json files to verify outputs:
+    ```bash
+    > tap-google-sheets --config tap_config.json --catalog catalog.json | target-json > state.json
+    > tail -1 state.json > state.json.tmp && mv state.json.tmp state.json
+    ```
+    To pseudo-load to [Stitch Import API](https://github.com/singer-io/target-stitch) with dry run:
+    ```bash
+    > tap-google-sheets --config tap_config.json --catalog catalog.json | target-stitch --config target_config.json --dry-run > state.json
+    > tail -1 state.json > state.json.tmp && mv state.json.tmp state.json
+    ```
+
+6. Test the Tap
+    
+    While developing the Google Search Console tap, the following utilities were run in accordance with Singer.io best practices:
+    Pylint to improve [code quality](https://github.com/singer-io/getting-started/blob/master/docs/BEST_PRACTICES.md#code-quality):
+    ```bash
+    > pylint tap_google_sheets -d missing-docstring -d logging-format-interpolation -d too-many-locals -d too-many-arguments
+    ```
+    Pylint test resulted in the following score:
+    ```bash
+    TBD
+    ```
+
+    To [check the tap](https://github.com/singer-io/singer-tools#singer-check-tap) and verify working:
+    ```bash
+    > tap-google-sheets --config tap_config.json --catalog catalog.json | singer-check-tap > state.json
+    > tail -1 state.json > state.json.tmp && mv state.json.tmp state.json
+    ```
+    Check tap resulted in the following:
+    ```bash
+    TBD
+    ```
+---
+
+Copyright &copy; 2019 Stitch
diff --git a/config.json.example b/config.json.example
new file mode 100644 (file)
index 0000000..159aafb
--- /dev/null
@@ -0,0 +1,8 @@
+{
+    "client_id": "YOUR_CLIENT_ID",
+    "client_secret": "YOUR_CLIENT_SECRET",
+    "refresh_token": "YOUR_REFRESH_TOKEN",
+    "spreadsheet_id": "YOUR_GOOGLE_SPREADSHEET_ID",
+    "start_date": "2019-01-01T00:00:00Z",
+    "user_agent": "tap-google-search-console <api_user_email@example.com>"
+}
diff --git a/setup.py b/setup.py
new file mode 100644 (file)
index 0000000..e3c4f3e
--- /dev/null
+++ b/setup.py
@@ -0,0 +1,25 @@
+#!/usr/bin/env python
+
+from setuptools import setup, find_packages
+
+setup(name='tap-google-sheets',
+      version='0.0.1',
+      description='Singer.io tap for extracting data from the Google Sheets v4 API',
+      author='jeff.huth@bytecode.io',
+      classifiers=['Programming Language :: Python :: 3 :: Only'],
+      py_modules=['tap_google_sheets'],
+      install_requires=[
+          'backoff==1.8.0',
+          'requests==2.22.0',
+          'singer-python==5.8.1'
+      ],
+      entry_points='''
+          [console_scripts]
+          tap-google-sheets=tap_google_sheets:main
+      ''',
+      packages=find_packages(),
+      package_data={
+          'tap_google_sheets': [
+              'schemas/*.json'
+          ]
+      })
diff --git a/state.json.example b/state.json.example
new file mode 100644 (file)
index 0000000..c33cc8f
--- /dev/null
@@ -0,0 +1,6 @@
+{
+       "currently_syncing": "spreadsheets",
+       "bookmarks": {
+               "file": "2019-09-27T22:34:39.000000Z"
+       }
+}
diff --git a/tap_google_sheets/__init__.py b/tap_google_sheets/__init__.py
new file mode 100644 (file)
index 0000000..f97d4b8
--- /dev/null
@@ -0,0 +1,57 @@
+#!/usr/bin/env python3
+
+import sys
+import json
+import argparse
+import singer
+from singer import metadata, utils
+from tap_google_sheets.client import GoogleClient
+from tap_google_sheets.discover import discover
+from tap_google_sheets.sync import sync
+
+LOGGER = singer.get_logger()
+
+REQUIRED_CONFIG_KEYS = [
+    'client_id',
+    'client_secret',
+    'refresh_token',
+    'spreadsheet_id',
+    'start_date',
+    'user_agent'
+]
+
+def do_discover(client, spreadsheet_id):
+
+    LOGGER.info('Starting discover')
+    catalog = discover(client, spreadsheet_id)
+    json.dump(catalog.to_dict(), sys.stdout, indent=2)
+    LOGGER.info('Finished discover')
+
+
+@singer.utils.handle_top_exception(LOGGER)
+def main():
+
+    parsed_args = singer.utils.parse_args(REQUIRED_CONFIG_KEYS)
+
+    with GoogleClient(parsed_args.config['client_id'],
+                      parsed_args.config['client_secret'],
+                      parsed_args.config['refresh_token'],
+                      parsed_args.config['user_agent']) as client:
+
+        state = {}
+        if parsed_args.state:
+            state = parsed_args.state
+
+        config = parsed_args.config
+        spreadsheet_id = config.get('spreadsheet_id')
+
+        if parsed_args.discover:
+            do_discover(client, spreadsheet_id)
+        elif parsed_args.catalog:
+            sync(client=client,
+                 config=config,
+                 catalog=parsed_args.catalog,
+                 state=state)
+
+if __name__ == '__main__':
+    main()
diff --git a/tap_google_sheets/client.py b/tap_google_sheets/client.py
new file mode 100644 (file)
index 0000000..12f0811
--- /dev/null
@@ -0,0 +1,247 @@
+from datetime import datetime, timedelta
+import backoff
+import requests
+from collections import OrderedDict
+
+import singer
+from singer import metrics
+from singer import utils
+
+BASE_URL = 'https://www.googleapis.com'
+GOOGLE_TOKEN_URI = 'https://oauth2.googleapis.com/token'
+LOGGER = singer.get_logger()
+
+
+class Server5xxError(Exception):
+    pass
+
+
+class Server429Error(Exception):
+    pass
+
+
+class GoogleError(Exception):
+    pass
+
+
+class GoogleBadRequestError(GoogleError):
+    pass
+
+
+class GoogleUnauthorizedError(GoogleError):
+    pass
+
+
+class GooglePaymentRequiredError(GoogleError):
+    pass
+
+
+class GoogleNotFoundError(GoogleError):
+    pass
+
+
+class GoogleMethodNotAllowedError(GoogleError):
+    pass
+
+
+class GoogleConflictError(GoogleError):
+    pass
+
+
+class GoogleGoneError(GoogleError):
+    pass
+
+
+class GooglePreconditionFailedError(GoogleError):
+    pass
+
+
+class GoogleRequestEntityTooLargeError(GoogleError):
+    pass
+
+
+class GoogleRequestedRangeNotSatisfiableError(GoogleError):
+    pass
+
+
+class GoogleExpectationFailedError(GoogleError):
+    pass
+
+
+class GoogleForbiddenError(GoogleError):
+    pass
+
+
+class GoogleUnprocessableEntityError(GoogleError):
+    pass
+
+
+class GooglePreconditionRequiredError(GoogleError):
+    pass
+
+
+class GoogleInternalServiceError(GoogleError):
+    pass
+
+
+# Error Codes: https://developers.google.com/webmaster-tools/search-console-api-original/v3/errors
+ERROR_CODE_EXCEPTION_MAPPING = {
+    400: GoogleBadRequestError,
+    401: GoogleUnauthorizedError,
+    402: GooglePaymentRequiredError,
+    403: GoogleForbiddenError,
+    404: GoogleNotFoundError,
+    405: GoogleMethodNotAllowedError,
+    409: GoogleConflictError,
+    410: GoogleGoneError,
+    412: GooglePreconditionFailedError,
+    413: GoogleRequestEntityTooLargeError,
+    416: GoogleRequestedRangeNotSatisfiableError,
+    417: GoogleExpectationFailedError,
+    422: GoogleUnprocessableEntityError,
+    428: GooglePreconditionRequiredError,
+    500: GoogleInternalServiceError}
+
+
+def get_exception_for_error_code(error_code):
+    return ERROR_CODE_EXCEPTION_MAPPING.get(error_code, GoogleError)
+
+def raise_for_error(response):
+    try:
+        response.raise_for_status()
+    except (requests.HTTPError, requests.ConnectionError) as error:
+        try:
+            content_length = len(response.content)
+            if content_length == 0:
+                # There is nothing we can do here since Google has neither sent
+                # us a 2xx response nor a response content.
+                return
+            response = response.json()
+            if ('error' in response) or ('errorCode' in response):
+                message = '%s: %s' % (response.get('error', str(error)),
+                                      response.get('message', 'Unknown Error'))
+                error_code = response.get('error', {}).get('code')
+                ex = get_exception_for_error_code(error_code)
+                raise ex(message)
+            else:
+                raise GoogleError(error)
+        except (ValueError, TypeError):
+            raise GoogleError(error)
+
+class GoogleClient: # pylint: disable=too-many-instance-attributes
+    def __init__(self,
+                 client_id,
+                 client_secret,
+                 refresh_token,
+                 user_agent=None):
+        self.__client_id = client_id
+        self.__client_secret = client_secret
+        self.__refresh_token = refresh_token
+        self.__user_agent = user_agent
+        self.__access_token = None
+        self.__expires = None
+        self.__session = requests.Session()
+        self.base_url = None
+
+
+    def __enter__(self):
+        self.get_access_token()
+        return self
+
+    def __exit__(self, exception_type, exception_value, traceback):
+        self.__session.close()
+
+    @backoff.on_exception(backoff.expo,
+                          Server5xxError,
+                          max_tries=5,
+                          factor=2)
+    def get_access_token(self):
+        # The refresh_token never expires and may be used many times to generate each access_token
+        # Since the refresh_token does not expire, it is not included in get access_token response
+        if self.__access_token is not None and self.__expires > datetime.utcnow():
+            return
+
+        headers = {}
+        if self.__user_agent:
+            headers['User-Agent'] = self.__user_agent
+
+        response = self.__session.post(
+            url=GOOGLE_TOKEN_URI,
+            headers=headers,
+            data={
+                'grant_type': 'refresh_token',
+                'client_id': self.__client_id,
+                'client_secret': self.__client_secret,
+                'refresh_token': self.__refresh_token,
+            })
+
+        if response.status_code >= 500:
+            raise Server5xxError()
+
+        if response.status_code != 200:
+            raise_for_error(response)
+
+        data = response.json()
+        self.__access_token = data['access_token']
+        self.__expires = datetime.utcnow() + timedelta(seconds=data['expires_in'])
+        LOGGER.info('Authorized, token expires = {}'.format(self.__expires))
+
+
+    @backoff.on_exception(backoff.expo,
+                          (Server5xxError, ConnectionError, Server429Error),
+                          max_tries=7,
+                          factor=3)
+    # Rate Limit:
+    #  https://developers.google.com/webmaster-tools/search-console-api-original/v3/limits
+    @utils.ratelimit(1200, 60)
+    def request(self, method, path=None, url=None, api=None, **kwargs):
+
+        self.get_access_token()
+        
+        self.base_url = 'https://sheets.googleapis.com/v4'
+        if api == 'files':
+            self.base_url = 'https://www.googleapis.com/drive/v3'
+
+        if not url and path:
+            url = '{}/{}'.format(self.base_url, path)
+
+        # endpoint = stream_name (from sync.py API call)
+        if 'endpoint' in kwargs:
+            endpoint = kwargs['endpoint']
+            del kwargs['endpoint']
+        else:
+            endpoint = None
+
+        if 'headers' not in kwargs:
+            kwargs['headers'] = {}
+        kwargs['headers']['Authorization'] = 'Bearer {}'.format(self.__access_token)
+
+        if self.__user_agent:
+            kwargs['headers']['User-Agent'] = self.__user_agent
+
+        if method == 'POST':
+            kwargs['headers']['Content-Type'] = 'application/json'
+
+        with metrics.http_request_timer(endpoint) as timer:
+            response = self.__session.request(method, url, **kwargs)
+            timer.tags[metrics.Tag.http_status_code] = response.status_code
+
+        if response.status_code >= 500:
+            raise Server5xxError()
+
+        #Use retry functionality in backoff to wait and retry if
+        #response code equals 429 because rate limit has been exceeded
+        if response.status_code == 429:
+            raise Server429Error()
+
+        if response.status_code != 200:
+            raise_for_error(response)
+
+        # Ensure keys and rows are ordered as received from API
+        return response.json(object_pairs_hook=OrderedDict)
+
+    def get(self, path, api, **kwargs):
+        return self.request(method='GET', path=path, api=api, **kwargs)
+
+    def post(self, path, api, **kwargs):
+        return self.request(method='POST', path=path, api=api, **kwargs)
diff --git a/tap_google_sheets/discover.py b/tap_google_sheets/discover.py
new file mode 100644 (file)
index 0000000..6477a5f
--- /dev/null
@@ -0,0 +1,26 @@
+from singer.catalog import Catalog, CatalogEntry, Schema
+from tap_google_sheets.schema import get_schemas, STREAMS
+
+
+def discover(client, spreadsheet_id):
+    schemas, field_metadata = get_schemas(client, spreadsheet_id)
+    catalog = Catalog([])
+
+    for stream_name, schema_dict in schemas.items():
+        schema = Schema.from_dict(schema_dict)
+        mdata = field_metadata[stream_name]
+        key_properties = None
+        for md in mdata:
+            table_key_properties = md.get('metadata', {}).get('table-key-properties')
+            if table_key_properties:
+                key_properties = table_key_properties
+    
+        catalog.streams.append(CatalogEntry(
+            stream=stream_name,
+            tap_stream_id=stream_name,
+            key_properties=STREAMS.get(stream_name, {}).get('key_properties', key_properties),
+            schema=schema,
+            metadata=mdata
+        ))
+
+    return catalog
diff --git a/tap_google_sheets/schema.py b/tap_google_sheets/schema.py
new file mode 100644 (file)
index 0000000..237ab06
--- /dev/null
@@ -0,0 +1,228 @@
+import os
+import json
+from collections import OrderedDict
+import singer
+from singer import metadata
+from tap_google_sheets.streams import STREAMS
+
+LOGGER = singer.get_logger()
+
+# Reference:
+# https://github.com/singer-io/getting-started/blob/master/docs/DISCOVERY_MODE.md#Metadata
+
+# Convert column index to column letter
+def colnum_string(n):
+    string = ""
+    while n > 0:
+        n, remainder = divmod(n - 1, 26)
+        string = chr(65 + remainder) + string
+    return string
+
+
+# Create sheet_metadata_json with columns from sheet
+def get_sheet_schema_columns(sheet, spreadsheet_id, client):
+    sheet_json_schema = OrderedDict()
+    data = next(iter(sheet.get('data', [])), {})
+    row_data = data.get('rowData',[])
+    # spreadsheet is an OrderedDict, with orderd sheets and rows in the repsonse
+
+    headers = row_data[0].get('values', [])
+    first_values = row_data[1].get('values', [])
+    # LOGGER.info('first_values = {}'.format(json.dumps(first_values, indent=2, sort_keys=True)))
+
+    sheet_json_schema['type'] = 'object'
+    sheet_json_schema['additionalProperties'] = False
+    sheet_json_schema = {
+        'type': 'object',
+        'additionalProperties': False,
+        'properties': {
+            '__sdc_spreadsheet_id': {
+                'type': ['null', 'string']
+            },
+            '__sdc_sheet_id': {
+                'type': ['null', 'integer']
+            },
+            '__sdc_row': {
+                'type': ['null', 'integer']
+            }
+        }
+    }
+
+    header_list = [] # used for checking uniqueness
+    columns = []
+    prior_header = None
+    i = 0
+    skipped = 0
+    # Read column headers until end or 2 consecutive skipped headers
+    for header in headers:
+        # LOGGER.info('header = {}'.format(json.dumps(header, indent=2, sort_keys=True)))
+        column_index = i + 1
+        column_letter = colnum_string(column_index)
+        header_value = header.get('formattedValue')
+        if header_value: # NOT skipped
+            column_is_skipped = False
+            skipped = 0
+            column_name = '{}'.format(header_value)
+            if column_name in header_list:
+                raise Exception('DUPLICATE HEADER ERROR: {}'.format(column_name))
+            else:
+                header_list.append(column_name)
+
+            first_value = first_values[i]
+            # LOGGER.info('first_value[{}] = {}'.format(i, json.dumps(first_value, indent=2, sort_keys=True)))
+
+            column_effective_value = first_value.get('effectiveValue', {})
+            for key in column_effective_value.keys():
+                if key in ('numberValue', 'stringValue', 'boolValue', 'errorType', 'formulaType'):
+                    column_effective_value_type = key
+
+            column_number_format = first_values[i].get('effectiveFormat', {}).get('numberFormat', {})
+            column_number_format_type = column_number_format.get('type')
+
+            # Determine datatype for sheet_json_schema
+            #
+            # column_effective_value_type = numberValue, stringValue, boolValue; INVALID: errorType, formulaType
+            #   Reference: https://developers.google.com/sheets/api/reference/rest/v4/spreadsheets/other#ExtendedValue
+            #
+            # column_number_format_type = UNEPECIFIED, TEXT, NUMBER, PERCENT, CURRENCY, DATE, TIME, DATE_TIME, SCIENTIFIC
+            #   Reference: https://developers.google.com/sheets/api/reference/rest/v4/spreadsheets/cells#NumberFormatType
+            #
+            column_format = None # Default
+            # column_multiple_of = None # Default
+            if column_effective_value_type in ('formulaValue', 'errorValue'):
+                raise Exception('INVALID DATA TYPE ERROR: {}, value: {}'.format(column_name))
+            elif column_effective_value_type == 'stringValue':
+                column_type = ['null', 'string']
+                column_gs_type = 'stringValue'
+            elif column_effective_value_type == 'boolValue':
+                column_type = ['null', 'boolean', 'string']
+                column_gs_type = 'boolValue'
+            elif column_effective_value_type == 'numberValue':
+                if column_number_format_type == 'DATE_TIME':
+                    column_type = ['null', 'string']
+                    column_format = 'date-time'
+                    column_gs_type = 'numberType.DATE_TIME'
+                elif column_number_format_type == 'DATE':
+                    column_type = ['null', 'string']
+                    column_format = 'date'
+                    column_gs_type = 'numberType.DATE'
+                elif column_number_format_type == 'TIME':
+                    column_type = ['null', 'string']
+                    column_format = 'time'
+                    column_gs_type = 'numberType.TIME'
+                elif column_number_format_type == 'TEXT':
+                    column_type = ['null', 'string']
+                    column_gs_type = 'stringValue'
+                else:
+                    column_type = ['null', 'number', 'string']
+                    column_gs_type = 'numberType'
+
+        else: # skipped
+            column_is_skipped = True
+            skipped = skipped + 1
+            column_index_str = str(column_index).zfill(2)
+            column_name = '__sdc_skip_col_{}'.format(column_index_str)
+            column_type = ['null', 'string']
+            column_format = None
+            column_gs_type = 'stringValue'
+
+        if skipped >= 2:
+            # skipped = 2 consecutive skipped headers
+            # Remove prior_header column_name
+            sheet_json_schema['properties'].pop(prior_header, None)
+            column_count = i - 1
+            break
+
+        else:
+            column = {}
+            column = {
+                'columnIndex': column_index,
+                'columnLetter': column_letter,
+                'columnName': column_name,
+                'columnType': column_gs_type,
+                'columnSkipped': column_is_skipped
+            }
+            columns.append(column)
+
+            sheet_json_schema['properties'][column_name] = column
+            sheet_json_schema['properties'][column_name]['type'] = column_type
+            if column_format:
+                sheet_json_schema['properties'][column_name]['format'] = column_format
+
+        prior_header = column_name
+        i = i + 1
+
+    return sheet_json_schema, columns
+
+
+def get_sheet_metadata(sheet, spreadsheet_id, client):
+    sheet_id = sheet.get('properties', {}).get('sheetId')
+    sheet_title = sheet.get('properties', {}).get('title')
+    LOGGER.info('sheet_id = {}, sheet_title = {}'.format(sheet_id, sheet_title))
+
+    stream_name = 'sheet_metadata'
+    stream_metadata = STREAMS.get(stream_name)
+    api = stream_metadata.get('api', 'sheets')
+    params = stream_metadata.get('params', {})
+    querystring = '&'.join(['%s=%s' % (key, value) for (key, value) in params.items()]).replace('{sheet_title}', sheet_title)
+    path = '{}?{}'.format(stream_metadata.get('path').replace('{spreadsheet_id}', spreadsheet_id), querystring)
+
+    sheet_md_results = client.get(path=path, api=api, endpoint=stream_name)
+    sheet_cols = sheet_md_results.get('sheets')[0]
+    sheet_schema, columns = get_sheet_schema_columns(sheet_cols, spreadsheet_id, client)
+
+    return sheet_schema, columns
+
+
+def get_abs_path(path):
+    return os.path.join(os.path.dirname(os.path.realpath(__file__)), path)
+
+def get_schemas(client, spreadsheet_id):
+    schemas = {}
+    field_metadata = {}
+
+    for stream_name, stream_metadata in STREAMS.items():
+        schema_path = get_abs_path('schemas/{}.json'.format(stream_name))
+        with open(schema_path) as file:
+            schema = json.load(file)
+        schemas[stream_name] = schema
+        mdata = metadata.new()
+
+        # Documentation:
+        # https://github.com/singer-io/getting-started/blob/master/docs/DISCOVERY_MODE.md#singer-python-helper-functions
+        # Reference:
+        # https://github.com/singer-io/singer-python/blob/master/singer/metadata.py#L25-L44
+        mdata = metadata.get_standard_metadata(
+            schema=schema,
+            key_properties=stream_metadata.get('key_properties', None),
+            valid_replication_keys=stream_metadata.get('replication_keys', None),
+            replication_method=stream_metadata.get('replication_method', None)
+        )
+        field_metadata[stream_name] = mdata
+        
+        if stream_name == 'spreadsheet_metadata':
+            api = stream_metadata.get('api', 'sheets')
+            params = stream_metadata.get('params', {})
+            querystring = '&'.join(['%s=%s' % (key, value) for (key, value) in params.items()])
+            path = '{}?{}'.format(stream_metadata.get('path').replace('{spreadsheet_id}', spreadsheet_id), querystring)
+
+            spreadsheet_md_results = client.get(path=path, params=querystring, api=api, endpoint=stream_name)
+
+            sheets = spreadsheet_md_results.get('sheets')
+            if sheets:
+                for sheet in sheets:
+                    sheet_schema, columns = get_sheet_metadata(sheet, spreadsheet_id, client)
+                    # LOGGER.info('sheet_schema = {}'.format(json.dumps(sheet_schema, indent=2, sort_keys=True)))
+
+                    sheet_title = sheet.get('properties', {}).get('title')
+                    schemas[sheet_title] = sheet_schema
+                    sheet_mdata = metadata.new()
+                    sheet_mdata = metadata.get_standard_metadata(
+                        schema=sheet_schema,
+                        key_properties=['__sdc_row'],
+                        valid_replication_keys=None,
+                        replication_method='FULL_TABLE'
+                    )
+                    field_metadata[sheet_title] = sheet_mdata
+            
+    return schemas, field_metadata
diff --git a/tap_google_sheets/schemas/file_metadata.json b/tap_google_sheets/schemas/file_metadata.json
new file mode 100644 (file)
index 0000000..25c19c4
--- /dev/null
@@ -0,0 +1,44 @@
+{
+  "type": "object",
+  "additionalProperties": false,
+  "properties": {
+    "id": {
+      "type": ["null", "string"]
+    },
+    "name": {
+      "type": ["null", "string"]
+    },
+    "version": {
+      "type": ["null", "integer"]
+    },
+    "createdTime": {
+      "type": ["null", "string"],
+      "format": "date-time"
+    },
+    "modifiedTime": {
+      "type": ["null", "string"],
+      "format": "date-time"
+    },
+    "teamDriveId": {
+      "type": ["null", "string"]
+    },
+    "driveId": {
+      "type": ["null", "string"]
+    },
+    "lastModifyingUser": {
+      "type": ["null", "object"],
+      "additionalProperties": false,
+      "properties": {
+        "kind": {
+          "type": ["null", "integer"]
+        },
+        "displayName": {
+          "type": ["null", "string"]
+        },
+        "emailAdress": {
+          "type": ["null", "string"]
+        }
+      }
+    }
+  }
+}
diff --git a/tap_google_sheets/schemas/sheet_metadata.json b/tap_google_sheets/schemas/sheet_metadata.json
new file mode 100644 (file)
index 0000000..c3f2ac2
--- /dev/null
@@ -0,0 +1,89 @@
+{
+  "type": "object",
+  "additionalProperties": false,
+  "properties": {
+    "spreadsheetId": {
+      "type": ["null", "string"]
+    },
+    "sheetId": {
+      "type": ["null", "integer"]
+    },
+    "title": {
+      "type": ["null", "string"]
+    },
+    "index": {
+      "type": ["null", "integer"]
+    },
+    "sheetType": {
+      "type": ["null", "string"]
+    },
+    "sheetUrl": {
+      "type": ["null", "string"]
+    },
+    "gridProperties": {
+      "type": ["null", "object"],
+      "additionalProperties": false,
+      "properties": {
+        "rowCount": {
+          "type": ["null", "integer"]
+        },
+        "columnCount": {
+          "type": ["null", "integer"]
+        },
+        "frozenRowCount": {
+          "type": ["null", "integer"]
+        },
+        "frozenColumnCount": {
+          "type": ["null", "integer"]
+        }
+      }
+    },
+    "columns": {
+      "anyOf": [
+        {
+          "type": "array",
+          "items": {
+            "type": ["null", "object"],
+            "additionalProperties": false,
+            "properties": {
+              "columnIndex": {
+                "type": ["null", "integer"]
+              },
+              "columnLetter": {
+                "type": ["null", "string"]
+              },
+              "columnName": {
+                "type": ["null", "string"]
+              },
+              "columnType": {
+                "type": ["null", "string"]
+              },
+              "columnSkipped": {
+                "type": ["null", "boolean"]
+              },
+              "type": {
+                "anyOf": [
+                  {
+                    "type": "array",
+                    "items": {
+                      "type": "string"
+                    }
+                  },
+                  {
+                    "type": "null"
+                  }
+                ]
+              },
+              "format": {
+                "type": ["null", "string"]
+              }
+            }
+          }
+        },
+        {
+          "type": "null"
+        }
+      ]
+    }
+  }
+}
diff --git a/tap_google_sheets/schemas/sheets_loaded.json b/tap_google_sheets/schemas/sheets_loaded.json
new file mode 100644 (file)
index 0000000..12f967a
--- /dev/null
@@ -0,0 +1,22 @@
+{
+  "type": "object",
+  "additionalProperties": false,
+  "properties": {
+    "spreadsheetId": {
+      "type": ["null", "string"]
+    },
+    "sheetId": {
+      "type": ["null", "integer"]
+    },
+    "sheetTitle": {
+      "type": ["null", "string"]
+    },
+    "loadDate": {
+      "type": ["null", "string"],
+      "format": "date-time"
+    },
+    "lastRowNumber": {
+      "type": ["null", "integer"]
+    }
+  }
+}
diff --git a/tap_google_sheets/schemas/spreadsheet_metadata.json b/tap_google_sheets/schemas/spreadsheet_metadata.json
new file mode 100644 (file)
index 0000000..852cb76
--- /dev/null
@@ -0,0 +1,30 @@
+{
+  "type": "object",
+  "additionalProperties": false,
+  "properties": {
+    "spreadsheetId": {
+      "type": ["null", "string"]
+    },
+    "properties": {
+      "type": ["null", "object"],
+      "additionalProperties": false,
+      "properties": {
+        "title": {
+          "type": ["null", "string"]
+        },
+        "locale": {
+          "type": ["null", "string"]
+        },
+        "autoRecalc": {
+          "type": ["null", "string"]
+        },
+        "timeZone": {
+          "type": ["null", "string"]
+        }
+      }
+    },
+    "spreadsheetUrl": {
+      "type": ["null", "string"]
+    }
+  }
+}
diff --git a/tap_google_sheets/streams.py b/tap_google_sheets/streams.py
new file mode 100644 (file)
index 0000000..299326a
--- /dev/null
@@ -0,0 +1,66 @@
+from collections import OrderedDict
+
+# streams: API URL endpoints to be called
+# properties:
+#   <root node>: Plural stream name for the endpoint
+#   path: API endpoint relative path, when added to the base URL, creates the full path,
+#       default = stream_name
+#   key_properties: Primary key fields for identifying an endpoint record.
+#   replication_method: INCREMENTAL or FULL_TABLE
+#   replication_keys: bookmark_field(s), typically a date-time, used for filtering the results
+#        and setting the state
+#   params: Query, sort, and other endpoint specific parameters; default = {}
+#   data_key: JSON element containing the results list for the endpoint; default = root (no data_key)
+#   bookmark_query_field: From date-time field used for filtering the query
+#   bookmark_type: Data type for bookmark, integer or datetime
+
+FILE_METADATA = {
+    "api": "files",
+    "path": "files/{spreadsheet_id}",
+    "key_properties": ["id"],
+    "replication_method": "FULL_TABLE",
+    "params": {
+        "fields": "id,name,createdTime,modifiedTime,version,teamDriveId,driveId,lastModifyingUser"
+    }
+}
+
+SPREADSHEET_METADATA = {
+    "api": "sheets",
+    "path": "spreadsheets/{spreadsheet_id}",
+    "key_properties": ["spreadsheetId"],
+    "replication_method": "FULL_TABLE",
+    "params": {
+        "includeGridData": "false"
+    }
+}
+
+SHEET_METADATA = {
+    "api": "sheets",
+    "path": "spreadsheets/{spreadsheet_id}",
+    "key_properties": ["sheetId"],
+    "replication_method": "FULL_TABLE",
+    "params": {
+        "includeGridData": "true",
+        "ranges": "'{sheet_title}'!1:2"
+    }
+}
+
+SHEETS_LOADED = {
+    "api": "sheets",
+    "path": "spreadsheets/{spreadsheet_id}/values/'{sheet_title}'!{range_rows}",
+    "data_key": "values",
+    "key_properties": ["spreadsheetId", "sheetId", "loadDate"],
+    "replication_method": "FULL_TABLE",
+    "params": {
+        "dateTimeRenderOption": "SERIAL_NUMBER",
+        "valueRenderOption": "UNFORMATTED_VALUE",
+        "majorDimension": "ROWS"
+    }
+}
+
+# Ensure streams are ordered logically
+STREAMS = OrderedDict()
+STREAMS['file_metadata'] = FILE_METADATA
+STREAMS['spreadsheet_metadata'] = SPREADSHEET_METADATA
+STREAMS['sheet_metadata'] = SHEET_METADATA
+STREAMS['sheets_loaded'] = SHEETS_LOADED
diff --git a/tap_google_sheets/sync.py b/tap_google_sheets/sync.py
new file mode 100644 (file)
index 0000000..a8b02d0
--- /dev/null
@@ -0,0 +1,281 @@
+import time
+import math
+import singer
+import json
+from collections import OrderedDict
+from singer import metrics, metadata, Transformer, utils
+from singer.utils import strptime_to_utc, strftime
+from tap_google_sheets.transform import transform_json
+from tap_google_sheets.streams import STREAMS
+from tap_google_sheets.schema import get_sheet_metadata
+
+LOGGER = singer.get_logger()
+
+
+def write_schema(catalog, stream_name):
+    stream = catalog.get_stream(stream_name)
+    schema = stream.schema.to_dict()
+    try:
+        singer.write_schema(stream_name, schema, stream.key_properties)
+    except OSError as err:
+        LOGGER.info('OS Error writing schema for: {}'.format(stream_name))
+        raise err
+
+
+def write_record(stream_name, record, time_extracted):
+    try:
+        singer.messages.write_record(stream_name, record, time_extracted=time_extracted)
+    except OSError as err:
+        LOGGER.info('OS Error writing record for: {}'.format(stream_name))
+        LOGGER.info('record: {}'.format(record))
+        raise err
+
+
+def get_bookmark(state, stream, default):
+    if (state is None) or ('bookmarks' not in state):
+        return default
+    return (
+        state
+        .get('bookmarks', {})
+        .get(stream, default)
+    )
+
+
+def write_bookmark(state, stream, value):
+    if 'bookmarks' not in state:
+        state['bookmarks'] = {}
+    state['bookmarks'][stream] = value
+    LOGGER.info('Write state for stream: {}, value: {}'.format(stream, value))
+    singer.write_state(state)
+
+
+# def transform_datetime(this_dttm):
+def transform_datetime(this_dttm):
+    with Transformer() as transformer:
+        new_dttm = transformer._transform_datetime(this_dttm)
+    return new_dttm
+
+
+def process_records(catalog, #pylint: disable=too-many-branches
+                    stream_name,
+                    records,
+                    time_extracted,
+                    bookmark_field=None,
+                    bookmark_type=None,
+                    max_bookmark_value=None,
+                    last_datetime=None,
+                    last_integer=None,
+                    parent=None,
+                    parent_id=None):
+    stream = catalog.get_stream(stream_name)
+    schema = stream.schema.to_dict()
+    stream_metadata = metadata.to_map(stream.metadata)
+
+    with metrics.record_counter(stream_name) as counter:
+        for record in records:
+            # If child object, add parent_id to record
+            if parent_id and parent:
+                record[parent + '_id'] = parent_id
+
+            # Transform record for Singer.io
+            with Transformer() as transformer:
+                transformed_record = transformer.transform(
+                    record,
+                    schema,
+                    stream_metadata)
+                # Reset max_bookmark_value to new value if higher
+                if transformed_record.get(bookmark_field):
+                    if max_bookmark_value is None or \
+                        transformed_record[bookmark_field] > transform_datetime(max_bookmark_value):
+                        max_bookmark_value = transformed_record[bookmark_field]
+
+                if bookmark_field and (bookmark_field in transformed_record):
+                    if bookmark_type == 'integer':
+                        # Keep only records whose bookmark is after the last_integer
+                        if transformed_record[bookmark_field] >= last_integer:
+                            write_record(stream_name, transformed_record, \
+                                time_extracted=time_extracted)
+                            counter.increment()
+                    elif bookmark_type == 'datetime':
+                        last_dttm = transform_datetime(last_datetime)
+                        bookmark_dttm = transform_datetime(transformed_record[bookmark_field])
+                        # Keep only records whose bookmark is after the last_datetime
+                        if bookmark_dttm >= last_dttm:
+                            write_record(stream_name, transformed_record, \
+                                time_extracted=time_extracted)
+                            counter.increment()
+                else:
+                    write_record(stream_name, transformed_record, time_extracted=time_extracted)
+                    counter.increment()
+
+        return max_bookmark_value, counter.value
+
+
+# Currently syncing sets the stream currently being delivered in the state.
+# If the integration is interrupted, this state property is used to identify
+#  the starting point to continue from.
+# Reference: https://github.com/singer-io/singer-python/blob/master/singer/bookmarks.py#L41-L46
+def update_currently_syncing(state, stream_name):
+    if (stream_name is None) and ('currently_syncing' in state):
+        del state['currently_syncing']
+    else:
+        singer.set_currently_syncing(state, stream_name)
+    singer.write_state(state)
+
+
+# List selected fields from stream catalog
+def get_selected_fields(catalog, stream_name):
+    stream = catalog.get_stream(stream_name)
+    mdata = metadata.to_map(stream.metadata)
+    mdata_list = singer.metadata.to_list(mdata)
+    selected_fields = []
+    for entry in mdata_list:
+        field =  None
+        try:
+            field =  entry['breadcrumb'][1]
+            if entry.get('metadata', {}).get('selected', False):
+                selected_fields.append(field)
+        except IndexError:
+            pass
+    return selected_fields
+
+
+def get_data(stream_name,
+             endpoint_config,
+             client,
+             spreadsheet_id,
+             range_rows=None):
+    if not range_rows:
+        range_rows = ''
+    path = endpoint_config.get('path', stream_name).replace(
+        '{spreadsheet_id}', spreadsheet_id).replace('{sheet_title}', stream_name).replace(
+            '{range_rows}', range_rows)
+    params = endpoint_config.get('params', {})
+    api = endpoint_config.get('api', 'sheets')
+    querystring = '&'.join(['%s=%s' % (key, value) for (key, value) in params.items()]).replace(
+        '{sheet_title}', stream_name)
+    data = {}
+    data = client.get(
+        path=path,
+        api=api,
+        params=querystring,
+        endpoint=stream_name)
+    return data
+
+
+def transform_file_metadata(file_metadata):
+    # Convert to dict
+    file_metadata_tf = json.loads(json.dumps(file_metadata))
+    # Remove keys
+    if file_metadata_tf.get('lastModifyingUser'):
+        file_metadata_tf['lastModifyingUser'].pop('photoLink', None)
+        file_metadata_tf['lastModifyingUser'].pop('me', None)
+        file_metadata_tf['lastModifyingUser'].pop('permissionId', None)
+    # Add record to an array of 1
+    file_metadata_arr = []
+    file_metadata_arr.append(file_metadata_tf)
+    return file_metadata_arr
+
+
+def transform_spreadsheet_metadata(spreadsheet_metadata):
+    # Convert to dict
+    spreadsheet_metadata_tf = json.loads(json.dumps(spreadsheet_metadata))
+    # Remove keys
+    if spreadsheet_metadata_tf.get('properties'):
+        spreadsheet_metadata_tf['properties'].pop('defaultFormat', None)
+    spreadsheet_metadata_tf.pop('sheets', None)
+    # Add record to an array of 1
+    spreadsheet_metadata_arr = []
+    spreadsheet_metadata_arr.append(spreadsheet_metadata_tf)
+    return spreadsheet_metadata_arr
+
+
+def transform_sheet_metadata(spreadsheet_id, sheet, columns):
+    # Convert to properties to dict
+    sheet_metadata = sheet.get('properties')
+    sheet_metadata_tf = json.loads(json.dumps(sheet_metadata)) 
+    sheet_id = sheet_metadata_tf.get('sheetId')
+    sheet_url = 'https://docs.google.com/spreadsheets/d/{}/edit#gid={}'.format(
+        spreadsheet_id, sheet_id)
+    sheet_metadata_tf['spreadsheetId'] = spreadsheet_id
+    sheet_metadata_tf['sheetUrl'] = sheet_url
+    sheet_metadata_tf['columns'] = columns
+    return sheet_metadata_tf
+
+
+def sync(client, config, catalog, state):
+    start_date = config.get('start_date')
+    spreadsheet_id = config.get('spreadsheet_id')
+
+    # Get selected_streams from catalog, based on state last_stream
+    #   last_stream = Previous currently synced stream, if the load was interrupted
+    last_stream = singer.get_currently_syncing(state)
+    LOGGER.info('last/currently syncing stream: {}'.format(last_stream))
+    selected_streams = []
+    for stream in catalog.get_selected_streams(state):
+        selected_streams.append(stream.stream)
+    LOGGER.info('selected_streams: {}'.format(selected_streams))
+
+    if not selected_streams:
+        return
+
+    # Get file_metadata
+    file_metadata = {}
+    file_metadata_config = STREAMS.get('file_metadata')
+    file_metadata = get_data('file_metadata', file_metadata_config, client, spreadsheet_id)
+    file_metadata_tf = transform_file_metadata(file_metadata)
+    # LOGGER.info('file_metadata_tf = {}'.format(file_metadata_tf))
+    last_datetime = strptime_to_utc(get_bookmark(state, 'file_metadata', start_date))
+    this_datetime = strptime_to_utc(file_metadata.get('modifiedTime'))
+    LOGGER.info('last_datetime = {}, this_datetime = {}'.format(last_datetime, this_datetime))
+    if this_datetime <= last_datetime:
+        LOGGER.info('this_datetime <= last_datetime, FILE NOT CHANGED. EXITING.')
+        return 0
+    
+    # Get spreadsheet_metadata
+    spreadsheet_metadata = {}
+    spreadsheet_metadata_config = STREAMS.get('spreadsheet_metadata')
+    spreadsheet_metadata = get_data('spreadsheet_metadata', spreadsheet_metadata_config, client, spreadsheet_id)
+    spreadsheet_metadata_tf = transform_spreadsheet_metadata(spreadsheet_metadata)
+    # LOGGER.info('spreadsheet_metadata_tf = {}'.format(spreadsheet_metadata_tf))
+
+    # Get sheet_metadata
+    sheets = spreadsheet_metadata.get('sheets')
+    sheet_metadata = []
+    sheets_loaded = []
+    sheets_loaded_config = STREAMS['sheets_loaded']
+    if sheets:
+        for sheet in sheets:
+            sheet_title = sheet.get('properties', {}).get('title')
+            sheet_schema, columns = get_sheet_metadata(sheet, spreadsheet_id, client)
+            sheet_metadata_tf = transform_sheet_metadata(spreadsheet_id, sheet, columns)
+            # LOGGER.info('sheet_metadata_tf = {}'.format(sheet_metadata_tf))
+            sheet_metadata.append(sheet_metadata_tf)
+
+            # Determine range of rows and columns for "paging" through batch rows of data
+            sheet_last_col_index = 1
+            sheet_last_col_letter = 'A'
+            for col in columns:
+                col_index = col.get('columnIndex')
+                col_letter = col.get('columnLetter')
+                if col_index > sheet_last_col_index:
+                    sheet_last_col_index = col_index
+                    sheet_last_col_letter = col_letter
+            sheet_max_row = sheet.get('gridProperties', {}).get('rowCount')
+            is_empty_row = False
+            batch_rows = 200
+            from_row = 2
+            if sheet_max_row < batch_rows:
+                to_row = sheet_max_row
+            else:
+                to_row = batch_rows
+
+            while not is_empty_row and to_row <= sheet_max_row:
+                range_rows = 'A2:{}{}'.format(sheet_last_col_letter, to_row)
+                
+                sheet_data = get_data(
+                    stream_name=sheet_title,
+                    endpoint_config=sheets_loaded_config,
+                    client=client,
+                    spreadsheet_id=spreadsheet_id,
+                    range_rows=range_rows)