Unpacking Software Livestream

Join our monthly Unpacking Software livestream to hear about the latest news, chat and opinion on packaging, software deployment and lifecycle management!

Learn More

Chocolatey Product Spotlight

Join the Chocolatey Team on our regular monthly stream where we put a spotlight on the most recent Chocolatey product releases. You'll have a chance to have your questions answered in a live Ask Me Anything format.

Learn More

Chocolatey Coding Livestream

Join us for the Chocolatey Coding Livestream, where members of our team dive into the heart of open source development by coding live on various Chocolatey projects. Tune in to witness real-time coding, ask questions, and gain insights into the world of package management. Don't miss this opportunity to engage with our team and contribute to the future of Chocolatey!

Learn More

Calling All Chocolatiers! Whipping Up Windows Automation with Chocolatey Central Management

Webinar from
Wednesday, 17 January 2024

We are delighted to announce the release of Chocolatey Central Management v0.12.0, featuring seamless Deployment Plan creation, time-saving duplications, insightful Group Details, an upgraded Dashboard, bug fixes, user interface polishing, and refined documentation. As an added bonus we'll have members of our Solutions Engineering team on-hand to dive into some interesting ways you can leverage the new features available!

Watch On-Demand
Chocolatey Community Coffee Break

Join the Chocolatey Team as we discuss all things Community, what we do, how you can get involved and answer your Chocolatey questions.

Watch The Replays
Chocolatey and Intune Overview

Webinar Replay from
Wednesday, 30 March 2022

At Chocolatey Software we strive for simple, and teaching others. Let us teach you just how simple it could be to keep your 3rd party applications updated across your devices, all with Intune!

Watch On-Demand
Chocolatey For Business. In Azure. In One Click.

Livestream from
Thursday, 9 June 2022

Join James and Josh to show you how you can get the Chocolatey For Business recommended infrastructure and workflow, created, in Azure, in around 20 minutes.

Watch On-Demand
The Future of Chocolatey CLI

Livestream from
Thursday, 04 August 2022

Join Paul and Gary to hear more about the plans for the Chocolatey CLI in the not so distant future. We'll talk about some cool new features, long term asks from Customers and Community and how you can get involved!

Watch On-Demand
Hacktoberfest Tuesdays 2022

Livestreams from
October 2022

For Hacktoberfest, Chocolatey ran a livestream every Tuesday! Re-watch Cory, James, Gary, and Rain as they share knowledge on how to contribute to open-source projects such as Chocolatey CLI.

Watch On-Demand



Downloads of v 4.3.2:


Last Update:

12 Apr 2021

Package Maintainer(s):

Software Author(s):

  • Arun Prakash Jana
  • Zhiming Wang


google search news cli command-line


  • 1
  • 2
  • 3

4.3.2 | Updated: 12 Apr 2021



Downloads of v 4.3.2:



Software Author(s):

  • Arun Prakash Jana
  • Zhiming Wang

googler 4.3.2

  • 1
  • 2
  • 3

All Checks are Passing

3 Passing Tests

Validation Testing Passed

Verification Testing Passed


Scan Testing Successful:

No detections found in any package files

Learn More

Deployment Method: Individual Install, Upgrade, & Uninstall

To install googler, run the following command from the command line or from PowerShell:


To upgrade googler, run the following command from the command line or from PowerShell:


To uninstall googler, run the following command from the command line or from PowerShell:


Deployment Method:


This applies to both open source and commercial editions of Chocolatey.

1. Enter Your Internal Repository Url

(this should look similar to https://community.chocolatey.org/api/v2/)

2. Setup Your Environment

1. Ensure you are set for organizational deployment

Please see the organizational deployment guide

2. Get the package into your environment

  • Open Source or Commercial:
    • Proxy Repository - Create a proxy nuget repository on Nexus, Artifactory Pro, or a proxy Chocolatey repository on ProGet. Point your upstream to https://community.chocolatey.org/api/v2/. Packages cache on first access automatically. Make sure your choco clients are using your proxy repository as a source and NOT the default community repository. See source command for more information.
    • You can also just download the package and push it to a repository Download

3. Copy Your Script

choco upgrade googler -y --source="'INTERNAL REPO URL'" [other options]

See options you can pass to upgrade.

See best practices for scripting.

Add this to a PowerShell script or use a Batch script with tools and in places where you are calling directly to Chocolatey. If you are integrating, keep in mind enhanced exit codes.

If you do use a PowerShell script, use the following to ensure bad exit codes are shown as failures:

choco upgrade googler -y --source="'INTERNAL REPO URL'" 

Write-Verbose "Exit code was $exitCode"
$validExitCodes = @(0, 1605, 1614, 1641, 3010)
if ($validExitCodes -contains $exitCode) {
  Exit 0

Exit $exitCode

- name: Install googler
    name: googler
    version: '4.3.2'
    state: present

See docs at https://docs.ansible.com/ansible/latest/modules/win_chocolatey_module.html.

chocolatey_package 'googler' do
  action    :install
  source   'INTERNAL REPO URL'
  version  '4.3.2'

See docs at https://docs.chef.io/resource_chocolatey_package.html.

cChocoPackageInstaller googler
    Name     = "googler"
    Version  = "4.3.2"
    Source   = "INTERNAL REPO URL"

Requires cChoco DSC Resource. See docs at https://github.com/chocolatey/cChoco.

package { 'googler':
  ensure   => '4.3.2',
  provider => 'chocolatey',
  source   => 'INTERNAL REPO URL',

Requires Puppet Chocolatey Provider module. See docs at https://forge.puppet.com/puppetlabs/chocolatey.

4. If applicable - Chocolatey configuration/installation

See infrastructure management matrix for Chocolatey configuration elements and examples.

Package Approved

This package was approved as a trusted package on 12 Jun 2024.


googler is a power tool to perform Google Search (Web or News) from the command-line. It shows the title, URL and abstract for each result, which can be directly opened in a browser from the terminal. Results are fetched in pages (with page navigation). Supports sequential searches in a single googler instance.

googler was initially written to cater to headless servers without X. You can integrate it with a text-based browser. However, it has grown into a very handy and flexible utility that delivers much more. For example, fetch any number of results or start anywhere, limit search by any duration, define aliases to google search any number of websites, switch domains easily... all of this in a very clean interface without ads or stray URLs. The shell completion scripts make sure you don't need to remember any options.

googler isn't affiliated to Google in any way.

                    GNU GENERAL PUBLIC LICENSE
                       Version 3, 29 June 2007

 Copyright (C) 2007 Free Software Foundation, Inc. <http://fsf.org/>
 Everyone is permitted to copy and distribute verbatim copies
 of this license document, but changing it is not allowed.


  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

  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

  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.


  16. Limitation of Liability.


  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 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
    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 <http://www.gnu.org/licenses/>.

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:

    <program>  Copyright (C) <year>  <name of author>
    This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
    This is free software, and you are welcome to redistribute it
    under certain conditions; type `show c' for details.

The hypothetical commands `show w' and `show c' should show the appropriate
parts of the General Public License.  Of course, your program's commands
might be different; for a GUI interface, you would use an "about box".

  You should also get your employer (if you work as a programmer) or school,
if any, to sign a "copyright disclaimer" for the program, if necessary.
For more information on this, and how to apply and follow the GNU GPL, see

  The GNU General Public License does not permit incorporating your program
into proprietary programs.  If your program is a subroutine library, you
may consider it more useful to permit linking proprietary applications with
the library.  If this is what you want to do, use the GNU Lesser General
Public License instead of this License.  But first, please read

Verification is intended to assist the Chocolatey moderators and community
in verifying that this package's contents are trustworthy.

Package can be verified like this:

1. Go to


   to download the script (renamed to 'googler.py' in the package).

2. You can use one of the following methods to obtain the SHA256 checksum:
   - Use powershell function 'Get-FileHash'
   - Use Chocolatey utility 'checksum.exe'

   checksum algorithm: SHA256
   checksum: 6819DE4B01F2683505892FB0DD1E06DC8C3B798279EC6684F5594BA4DC321E11

File 'LICENSE.txt' is obtained from:

Note that this package is maintained by one of the authors of googler.
$ErrorActionPreference = 'Stop';

$toolsDir = "$(Split-Path -parent $MyInvocation.MyCommand.Definition)"
Install-ChocolateyPath "$toolsDir\googler"
#!/usr/bin/env python3
# Copyright © 2008 Henri Hakkinen
# Copyright © 2015-2021 Arun Prakash Jana <[email protected]>
# 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
# 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 <http://www.gnu.org/licenses/>.

import argparse
import atexit
import base64
import collections
import codecs
import functools
import gzip
import html.entities
import html.parser
import http.client
from http.client import HTTPSConnection
import locale
import logging
import os
import platform
import shutil
import signal
import socket
import ssl
import subprocess
from subprocess import Popen, PIPE, DEVNULL
import sys
import textwrap
import unicodedata
import urllib.parse
import uuid
import webbrowser

# Python optional dependency compatibility layer
    import readline
except ImportError:

    import setproctitle
except (ImportError, Exception):

from typing import (

# Basic setup

logging.basicConfig(format='[%(levelname)s] %(message)s')
logger = logging.getLogger()

def sigint_handler(signum, frame):
    print('\nInterrupted.', file=sys.stderr)

    signal.signal(signal.SIGINT, sigint_handler)
except ValueError:
    # signal only works in main thread

# Constants

_VERSION_ = '4.3.2'
_EPOCH_ = '20210115'

COLORMAP = {k: '\x1b[%sm' % v for k, v in {
    'a': '30', 'b': '31', 'c': '32', 'd': '33',
    'e': '34', 'f': '35', 'g': '36', 'h': '37',
    'i': '90', 'j': '91', 'k': '92', 'l': '93',
    'm': '94', 'n': '95', 'o': '96', 'p': '97',
    'A': '30;1', 'B': '31;1', 'C': '32;1', 'D': '33;1',
    'E': '34;1', 'F': '35;1', 'G': '36;1', 'H': '37;1',
    'I': '90;1', 'J': '91;1', 'K': '92;1', 'L': '93;1',
    'M': '94;1', 'N': '95;1', 'O': '96;1', 'P': '97;1',
    'x': '0', 'X': '1', 'y': '7', 'Y': '7;1',

USER_AGENT = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36'

text_browsers = ['elinks', 'links', 'lynx', 'w3m', 'www-browser']

# Self-upgrade parameters
# Downstream packagers are recommended to turn off the entire self-upgrade
# mechanism through
#     make disable-self-upgrade
# before running `make install'.

API_REPO_BASE = 'https://api.github.com/repos/jarun/googler'
RAW_DOWNLOAD_REPO_BASE = 'https://raw.githubusercontent.com/jarun/googler'

debugger = False

# Monkeypatch textwrap for CJK wide characters.

def monkeypatch_textwrap_for_cjk():
        if textwrap.wrap.patched:
    except AttributeError:
    psl_textwrap_wrap = textwrap.wrap

    def textwrap_wrap(text, width=70, **kwargs):
        if width <= 2:
            width = 2
        # We first add a U+0000 after each East Asian Fullwidth or East
        # Asian Wide character, then fill to width - 1 (so that if a NUL
        # character ends up on a new line, we still have one last column
        # to spare for the preceding wide character). Finally we strip
        # all the NUL characters.
        # East Asian Width: https://www.unicode.org/reports/tr11/
        return [
            line.replace('\0', '')
            for line in psl_textwrap_wrap(
                    ch + '\0' if unicodedata.east_asian_width(ch) in ('F', 'W') else ch
                    for ch in unicodedata.normalize('NFC', text)
                width=width - 1,

    def textwrap_fill(text, width=70, **kwargs):
        return '\n'.join(textwrap_wrap(text, width=width, **kwargs))

    textwrap.wrap = textwrap_wrap
    textwrap.fill = textwrap_fill
    textwrap.wrap.patched = True
    textwrap.fill.patched = True


CoordinateType = Tuple[int, int]

class TrackedTextwrap:
    Implements a text wrapper that tracks the position of each source
    character, and can correctly insert zero-width sequences at given
    offsets of the source text.

    Wrapping result should be the same as that from PSL textwrap.wrap
    with default settings except expand_tabs=False.

    def __init__(self, text: str, width: int):
        self._original = text

        # Do the job of replace_whitespace first so that we can easily
        # match text to wrapped lines later. Note that this operation
        # does not change text length or offsets.
        whitespace = "\t\n\v\f\r "
        whitespace_trans = str.maketrans(whitespace, " " * len(whitespace))
        text = text.translate(whitespace_trans)

        self._lines = textwrap.wrap(
            text, width, expand_tabs=False, replace_whitespace=False

        # self._coords track the (row, column) coordinate of each source
        # character in the result text. It is indexed by offset in
        # source text.
        self._coords = []  # type: List[CoordinateType]
        offset = 0
            if not self._lines:
                # Source text only has whitespaces. We add an empty line
                # in order to produce meaningful coordinates.
                self._lines = [""]
            for row, line in enumerate(self._lines):
                assert text[offset : offset + len(line)] == line
                col = 0
                for _ in line:
                    self._coords.append((row, col))
                    offset += 1
                    col += 1
                # All subsequent dropped whitespaces map to the last, imaginary column
                # (the EOL character if you wish) of the current line.
                while offset < len(text) and text[offset] == " ":
                    self._coords.append((row, col))
                    offset += 1
            # One past the final character (think of it as EOF) should
            # be treated as a valid offset.
            self._coords.append((row, col))
        except AssertionError:
            raise RuntimeError(
                "TrackedTextwrap: the impossible happened at offset {} of text {!r}".format(
                    offset, self._original

    # seq should be a zero-width sequence, e.g., an ANSI escape sequence.
    # May raise IndexError if offset is out of bounds.
    def insert_zero_width_sequence(self, seq: str, offset: int) -> None:
        row, col = self._coords[offset]
        line = self._lines[row]
        self._lines[row] = line[:col] + seq + line[col:]

        # Shift coordinates of all characters after the given character
        # on the same line.
        shift = len(seq)
        offset += 1
        while offset < len(self._coords) and self._coords[offset][0] == row:
            _, col = self._coords[offset]
            self._coords[offset] = (row, col + shift)
            offset += 1

    def original(self) -> str:
        return self._original

    def lines(self) -> List[str]:
        return self._lines

    def wrapped(self) -> str:
        return "\n".join(self._lines)

    # May raise IndexError if offset is out of bounds.
    def get_coordinate(self, offset: int) -> CoordinateType:
        return self._coords[offset]

### begin dim (DOM implementation with CSS support) ###
### https://github.com/zmwangx/dim/blob/master/dim.py ###

import html
import re
from collections import OrderedDict
from enum import Enum
from html.parser import HTMLParser

SelectorGroupLike = Union[str, "SelectorGroup", "Selector"]

class Node(object):
    Represents a DOM node.

    Parts of JavaScript's DOM ``Node`` API and ``Element`` API are
    mirrored here, with extensions. In particular, ``querySelector`` and
    ``querySelectorAll`` are mirrored.

    Notable properties and methods: :meth:`attr()`, :attr:`classes`,
    :attr:`html`, :attr:`text`, :meth:`ancestors()`,
    :meth:`descendants()`, :meth:`select()`, :meth:`select_all()`,

        tag      (:class:`Optional`\\[:class:`str`])
        attrs    (:class:`Dict`\\[:class:`str`, :class:`str`])
        parent   (:class:`Optional`\\[:class:`Node`])
        children (:class:`List`\\[:class:`Node`])

    # Meant to be reimplemented by subclasses.
    def __init__(self) -> None:
        self.tag = None  # type: Optional[str]
        self.attrs = {}  # type: Dict[str, str]
        self.parent = None  # type: Optional[Node]
        self.children = []  # type: List[Node]

        # Used in DOMBuilder.
        self._partial = False
        self._namespace = None  # type: Optional[str]

    # HTML representation of the node. Meant to be implemented by
    # subclasses.
    def __str__(self) -> str:  # pragma: no cover
        raise NotImplementedError

    def select(self, selector: SelectorGroupLike) -> Optional["Node"]:
        """DOM ``querySelector`` clone. Returns one match (if any)."""
        selector = self._normalize_selector(selector)
        for node in self._select_all(selector):
            return node
        return None

    def query_selector(self, selector: SelectorGroupLike) -> Optional["Node"]:
        """Alias of :meth:`select`."""
        return self.select(selector)

    def select_all(self, selector: SelectorGroupLike) -> List["Node"]:
        """DOM ``querySelectorAll`` clone. Returns all matches in a list."""
        selector = self._normalize_selector(selector)
        return list(self._select_all(selector))

    def query_selector_all(self, selector: SelectorGroupLike) -> List["Node"]:
        """Alias of :meth:`select_all`."""
        return self.select_all(selector)

    def matched_by(
        self, selector: SelectorGroupLike, root: Optional["Node"] = None
    ) -> bool:
        Checks whether this node is matched by `selector`.

        See :meth:`SelectorGroup.matches()`.
        selector = self._normalize_selector(selector)
        return selector.matches(self, root=root)

    def _normalize_selector(selector: SelectorGroupLike) -> "SelectorGroup":
        if isinstance(selector, str):
            return SelectorGroup.from_str(selector)
        if isinstance(selector, SelectorGroup):
            return selector
        if isinstance(selector, Selector):
            return SelectorGroup([selector])
        raise ValueError("not a selector or group of selectors: %s" % repr(selector))

    def _select_all(self, selector: "SelectorGroup") -> Generator["Node", None, None]:
        for descendant in self.descendants():
            if selector.matches(descendant, root=self):
                yield descendant

    def child_nodes(self) -> List["Node"]:
        return self.children

    def first_child(self) -> Optional["Node"]:
        if self.children:
            return self.children[0]
            return None

    def first_element_child(self) -> Optional["Node"]:
        for child in self.children:
            if isinstance(child, ElementNode):
                return child
        return None

    def last_child(self) -> Optional["Node"]:
        if self.children:
            return self.children[-1]
            return None

    def last_element_child(self) -> Optional["Node"]:
        for child in reversed(self.children):
            if isinstance(child, ElementNode):
                return child
        return None

    def next_sibling(self) -> Optional["Node"]:
        """.. note:: Not O(1), use with caution."""
        next_siblings = self.next_siblings()
        if next_siblings:
            return next_siblings[0]
            return None

    def next_siblings(self) -> List["Node"]:
        parent = self.parent
        if not parent:
            return []
            index = parent.children.index(self)
            return parent.children[index + 1 :]
        except ValueError:  # pragma: no cover
            raise ValueError("node is not found in children of its parent")

    def next_element_sibling(self) -> Optional["ElementNode"]:
        """.. note:: Not O(1), use with caution."""
        for sibling in self.next_siblings():
            if isinstance(sibling, ElementNode):
                return sibling
        return None

    def previous_sibling(self) -> Optional["Node"]:
        """.. note:: Not O(1), use with caution."""
        previous_siblings = self.previous_siblings()
        if previous_siblings:
            return previous_siblings[0]
            return None

    def previous_siblings(self) -> List["Node"]:
        Compared to the natural DOM order, the order of returned nodes
        are reversed. That is, the adjacent sibling (if any) is the
        first in the returned list.
        parent = self.parent
        if not parent:
            return []
            index = parent.children.index(self)
            if index > 0:
                return parent.children[index - 1 :: -1]
                return []
        except ValueError:  # pragma: no cover
            raise ValueError("node is not found in children of its parent")

    def previous_element_sibling(self) -> Optional["ElementNode"]:
        """.. note:: Not O(1), use with caution."""
        for sibling in self.previous_siblings():
            if isinstance(sibling, ElementNode):
                return sibling
        return None

    def ancestors(
        self, *, root: Optional["Node"] = None
    ) -> Generator["Node", None, None]:
        Ancestors are generated in reverse order of depth, stopping at

        A :class:`RuntimeException` is raised if `root` is not in the
        ancestral chain.
        if self is root:
        ancestor = self.parent
        while ancestor is not root:
            if ancestor is None:
                raise RuntimeError("provided root node not found in ancestral chain")
            yield ancestor
            ancestor = ancestor.parent
        if root:
            yield root

    def descendants(self) -> Generator["Node", None, None]:
        """Descendants are generated in depth-first order."""
        for child in self.children:
            yield child
            yield from child.descendants()

    def attr(self, attr: str) -> Optional[str]:
        """Returns the attribute if it exists on the node, otherwise ``None``."""
        return self.attrs.get(attr)

    def html(self) -> str:
        HTML representation of the node.

        (For a :class:`TextNode`, :meth:`html` returns the escaped version of the
        return str(self)

    def outer_html(self) -> str:
        """Alias of :attr:`html`."""
        return self.html

    def inner_html(self) -> str:
        """HTML representation of the node's children."""
        return "".join(child.html for child in self.children)

    def text(self) -> str:  # pragma: no cover
        """This property is expected to be implemented by subclasses."""
        raise NotImplementedError

    def text_content(self) -> str:
        """Alias of :attr:`text`."""
        return self.text

    def classes(self) -> List[str]:
        return self.attrs.get("class", "").split()

    def class_list(self) -> List[str]:
        return self.classes

class ElementNode(Node):
    Represents an element node.

    Note that tag and attribute names are case-insensitive; attribute
    values are case-sensitive.

    def __init__(
        tag: str,
        attrs: Iterable[Tuple[str, Optional[str]]],
        parent: Optional["Node"] = None,
        children: Optional[Sequence["Node"]] = None
    ) -> None:
        self.tag = tag.lower()  # type: str
        self.attrs = OrderedDict((attr.lower(), val or "") for attr, val in attrs)
        self.parent = parent
        self.children = list(children or [])

    def __repr__(self) -> str:
        s = "<" + self.tag
        if self.attrs:
            s += " attrs=%s" % repr(list(self.attrs.items()))
        if self.children:
            s += " children=%s" % repr(self.children)
        s += ">"
        return s

    # https://ipython.readthedocs.io/en/stable/api/generated/IPython.lib.pretty.html
    def _repr_pretty_(self, p: Any, cycle: bool) -> None:  # pragma: no cover
        if cycle:
            raise RuntimeError("cycle detected in DOM tree")
        p.text("<\x1b[1m%s\x1b[0m" % self.tag)
        if self.attrs:
            p.text(" attrs=%s" % repr(list(self.attrs.items())))
        if self.children:
            p.text(" children=[")
            if len(self.children) == 1 and isinstance(self.first_child(), TextNode):
                p.text("\x1b[4m%s\x1b[0m" % repr(self.first_child()))
                with p.indent(2):
                    for child in self.children:
                        if hasattr(child, "_repr_pretty_"):
                            child._repr_pretty_(p, False)  # type: ignore
                            p.text("\x1b[4m%s\x1b[0m" % repr(child))

    def __str__(self) -> str:
        """HTML representation of the node."""
        s = "<" + self.tag
        for attr, val in self.attrs.items():
            s += ' %s="%s"' % (attr, html.escape(val))
        if self.children:
            s += ">"
            s += "".join(str(child) for child in self.children)
            s += "</%s>" % self.tag
            if _tag_is_void(self.tag):
                s += "/>"
                s += "></%s>" % self.tag
        return s

    def text(self) -> str:
        """The concatenation of all descendant text nodes."""
        return "".join(child.text for child in self.children)

class TextNode(str, Node):
    Represents a text node.

    Subclasses :class:`Node` and :class:`str`.

    def __new__(cls, text: str) -> "TextNode":
        s = str.__new__(cls, text)  # type: ignore
        s.parent = None
        return s  # type: ignore

    def __init__(self, text: str) -> None:

    def __repr__(self) -> str:
        return "<%s>" % str.__repr__(self)

    # HTML-escaped form of the text node. use text() for unescaped
    # version.
    def __str__(self) -> str:
        return html.escape(self)

    def __eq__(self, other: object) -> bool:
        Two text nodes are equal if and only if they are the same node.

        For string comparison, use :attr:`text`.
        return self is other

    def __ne__(self, other: object) -> bool:
        Two text nodes are non-equal if they are not the same node.

        For string comparison, use :attr:`text`.
        return self is not other

    def text(self) -> str:
        return str.__str__(self)

class DOMBuilderException(Exception):
    Exception raised when :class:`DOMBuilder` detects a bad state.

        pos (:class:`Tuple`\\[:class:`int`, :class:`int`]):
            Line number and offset in HTML input.
        why (:class:`str`):
            Reason of the exception.

    def __init__(self, pos: Tuple[int, int], why: str) -> None:
        self.pos = pos
        self.why = why

    def __str__(self) -> str:  # pragma: no cover
        return "DOM builder aborted at %d:%d: %s" % (self.pos[0], self.pos[1], self.why)

class DOMBuilder(HTMLParser):
    HTML parser / DOM builder.

    Subclasses :class:`html.parser.HTMLParser`.

    Consume HTML and builds a :class:`Node` tree. Once finished, use
    :attr:`root` to access the root of the tree.

    This parser cannot parse malformed HTML with tag mismatch.

    def __init__(self) -> None:
        # _stack is the stack for nodes. Each node is pushed to the
        # stack when its start tag is processed, and remains on the
        # stack until its parent node is completed (end tag processed),
        # at which point the node is attached to the parent node as a
        # child and popped from the stack.
        self._stack = []  # type: List[Node]
        # _namespace_stack is another stack tracking the parsing
        # context, which is generally the default namespace (None) but
        # changes when parsing foreign objects (e.g. 'svg' when parsing
        # an <svg>). The top element is always the current parsing
        # context, so popping works differently from _stack: an element
        # is popped as soon as the corresponding end tag is processed.
        self._namespace_stack = [None]  # type: List[Optional[str]]

    def handle_starttag(
        self, tag: str, attrs: Sequence[Tuple[str, Optional[str]]]
    ) -> None:
        node = ElementNode(tag, attrs)
        node._partial = True
        namespace = (
            if _tag_encloses_foreign_namespace(tag)
            else self._namespace_stack[-1]  # Inherit parent namespace
        node._namespace = namespace
        # For void elements (not in a foreign context), immediately
        # invoke the end tag handler (see handle_startendtag()).
        if not namespace and _tag_is_void(tag):

    def handle_endtag(self, tag: str) -> None:
        tag = tag.lower()
        children = []
        while self._stack and not self._stack[-1]._partial:
        if not self._stack:
            raise DOMBuilderException(self.getpos(), "extra end tag: %s" % repr(tag))
        parent = self._stack[-1]
        if parent.tag != tag:
            raise DOMBuilderException(
                "expecting end tag %s, got %s" % (repr(parent.tag), repr(tag)),
        parent.children = list(reversed(children))
        parent._partial = False
        for child in children:
            child.parent = parent

    # Make parser behavior for explicitly and implicitly void elements
    # (e.g., <hr> vs <hr/>) consistent. The former triggers
    # handle_starttag only, whereas the latter triggers
    # handle_startendtag (which by default triggers both handle_starttag
    # and handle_endtag). See https://bugs.python.org/issue25258.
    # An exception is foreign elements, which aren't considered void
    # elements but can be explicitly marked as self-closing according to
    # the HTML spec (e.g. <path/> is valid but <path> is not).
    # Therefore, both handle_starttag and handle_endtag must be called,
    # and handle_endtag should not be triggered from within
    # handle_starttag in that case.
    # Note that for simplicity we do not check whether the foreign
    # element in question is allowed to be self-closing by spec. (The
    # SVG spec unfortunately doesn't provide a readily available list of
    # such elements.)
    # https://html.spec.whatwg.org/multipage/syntax.html#foreign-elements
    def handle_startendtag(
        self, tag: str, attrs: Sequence[Tuple[str, Optional[str]]]
    ) -> None:
        if self._namespace_stack[-1] or _tag_encloses_foreign_namespace(tag):
            self.handle_starttag(tag, attrs)
            self.handle_starttag(tag, attrs)

    def handle_data(self, text: str) -> None:
        if not self._stack:
            # Ignore text nodes before the first tag.

    def root(self) -> "Node":
        Finishes processing and returns the root node.

        Raises :class:`DOMBuilderException` if there is no root tag or
        root tag is not closed yet.
        if not self._stack:
            raise DOMBuilderException(self.getpos(), "no root tag")
        if self._stack[0]._partial:
            raise DOMBuilderException(self.getpos(), "root tag not closed yet")
        return self._stack[0]

def parse_html(html: str, *, ParserClass: type = DOMBuilder) -> "Node":
    Parses HTML string, builds DOM, and returns root node.

    The parser may raise :class:`DOMBuilderException`.

        html: input HTML string
        ParserClass: :class:`DOMBuilder` or a subclass

        Root note of the parsed tree. If the HTML string contains
        multiple top-level elements, only the first is returned and the
        rest are lost.
    builder = ParserClass()  # type: DOMBuilder
    return builder.root

class SelectorParserException(Exception):
    Exception raised when the selector parser fails to parse an input.

        s (:class:`str`):
            The input string to be parsed.
        cursor (:class:`int`):
            Cursor position where the failure occurred.
        why (:class:`str`):
            Reason of the failure.

    def __init__(self, s: str, cursor: int, why: str) -> None:
        self.s = s
        self.cursor = cursor
        self.why = why

    def __str__(self) -> str:  # pragma: no cover
        return "selector parser aborted at character %d of %s: %s" % (

class SelectorGroup:
    Represents a group of CSS selectors.

    A group of CSS selectors is simply a comma-separated list of
    selectors. [#]_ See :class:`Selector` documentation for the scope of

    Typically, a :class:`SelectorGroup` is constructed from a string
    (e.g., ``th.center, td.center``) using the factory function

    .. [#] https://www.w3.org/TR/selectors-3/#grouping

    def __init__(self, selectors: Iterable["Selector"]) -> None:
        self._selectors = list(selectors)

    def __repr__(self) -> str:
        return "<SelectorGroup %s>" % repr(str(self))

    def __str__(self) -> str:
        return ", ".join(str(selector) for selector in self._selectors)

    def __len__(self) -> int:
        return len(self._selectors)

    def __getitem__(self, index: int) -> "Selector":
        return self._selectors[index]

    def __iter__(self) -> Iterator["Selector"]:
        return iter(self._selectors)

    def from_str(cls, s: str) -> "SelectorGroup":
        Parses input string into a group of selectors.

        :class:`SelectorParserException` is raised on invalid input. See
        :class:`Selector` documentation for the scope of support.

            s: input string

            Parsed group of selectors.
        i = 0
        selectors = []
        while i < len(s):
            selector, i = Selector.from_str(s, i)
        if not selectors:
            raise SelectorParserException(s, i, "selector group is empty")
        return cls(selectors)

    def matches(self, node: "Node", root: Optional["Node"] = None) -> bool:
        Decides whether the group of selectors matches `node`.

        The group of selectors matches `node` as long as one of the
        selectors matches `node`.

        If `root` is provided and child and/or descendant combinators
        are involved, parent/ancestor lookup terminates at `root`.
        return any(selector.matches(node, root=root) for selector in self)

class Selector:
    Represents a CSS selector.

    Recall that a CSS selector is a chain of one or more *sequences of
    simple selectors* separated by *combinators*. [#selectors-3]_ This
    concept is represented as a cons list of sequences of simple
    selectors (in right to left order). This class in fact holds a
    single sequence, with an optional combinator and reference to the
    previous sequence.

    For instance, ``main#main p.important.definition >
    a.term[id][href]`` would be parsed into (schematically) the
    following structure::

        ">" tag='a' classes=('term') attrs=([id], [href]) ~>
        " " tag='p' classes=('important', 'definition') ~>
        tag='main' id='main'

    Each line is held in a separate instance of :class:`Selector`,
    linked together by the :attr:`previous` attribute.

    Supported grammar (from selectors level 3 [#selectors-3]_):

    - Type selectors;
    - Universal selectors;
    - Class selectors;
    - ID selectors;
    - Attribute selectors;
    - Combinators.

    Unsupported grammar:

    - Pseudo-classes;
    - Pseudo-elements;
    - Namespace prefixes (``ns|``, ``*|``, ``|``) in any part of any


    - Pseudo-classes have too many variants, a few of which even
      complete with an admittedly not-so-complex minilanguage. These add
      up to a lot of code.
    - Pseudo-elements are useless outside rendering contexts, hence out of
    - Namespace support is too niche to be worth the parsing headache.
      *Using namespace prefixes may confuse the parser!*

    Note that the parser only loosely follows the spec and priotizes
    ease of parsing (which includes readability and *writability* of
    regexes), so some invalid selectors may be accepted (in fact, false
    positives abound, but accepting valid inputs is a much more
    important goal than rejecting invalid inputs for this library), and
    some valid selectors may be rejected (but as long as you stick to
    the scope outlined above and common sense you should be fine; the
    false negatives shouldn't be used by actual human beings anyway).

    In particular, whitespace character is simplified to ``\\s`` (ASCII
    mode) despite CSS spec not counting U+000B (VT) as whitespace,
    identifiers are simplified to ``[\\w-]+`` (ASCII mode), and strings
    (attribute selector values can be either identifiers or strings)
    allow escaped quotes (i.e., ``\\'`` inside single-quoted strings and
    ``\\"`` inside double-quoted strings) but everything else is
    interpreted literally. The exact specs for CSS identifiers and
    strings can be found at [#]_.

    Certain selectors and combinators may be implemented in the parser
    but not implemented in matching and/or selection APIs.

    .. [#selectors-3] https://www.w3.org/TR/selectors-3/
    .. [#] https://www.w3.org/TR/CSS21/syndata.html

        tag (:class:`Optional`\\[:class:`str`]):
            Type selector.
        classes (:class:`List`\\[:class:`str`]):
            Class selectors.
        id (:class:`Optional`\\[:class:`str`]):
            ID selector.
        attrs (:class:`List`\\[:class:`AttributeSelector`]):
            Attribute selectors.
        combinator (:class:`Optional`\\[:class:`Combinator`]):
            Combinator with the previous sequence of simple selectors in
        previous (:class:`Optional`\\[:class:`Selector`]):
            Reference to the previous sequence of simple selectors in


    def __init__(
        tag: Optional[str] = None,
        classes: Optional[Sequence[str]] = None,
        id: Optional[str] = None,
        attrs: Optional[Sequence["AttributeSelector"]] = None,
        combinator: Optional["Combinator"] = None,
        previous: Optional["Selector"] = None
    ) -> None:
        self.tag = tag.lower() if tag else None
        self.classes = list(classes or [])
        self.id = id
        self.attrs = list(attrs or [])
        self.combinator = combinator
        self.previous = previous

    def __repr__(self) -> str:
        return "<Selector %s>" % repr(str(self))

    def __str__(self) -> str:
        sequences = []
        delimiters = []
        seq = self
        while True:
            if seq.previous:
                if seq.combinator == Combinator.DESCENDANT:
                    delimiters.append(" ")
                elif seq.combinator == Combinator.CHILD:
                    delimiters.append(" > ")
                elif seq.combinator == Combinator.NEXT_SIBLING:
                    delimiters.append(" + ")
                elif seq.combinator == Combinator.SUBSEQUENT_SIBLING:
                    delimiters.append(" ~ ")
                else:  # pragma: no cover
                    raise RuntimeError(
                        "unimplemented combinator: %s" % repr(self.combinator)
                seq = seq.previous
        return "".join(
            delimiter + sequence
            for delimiter, sequence in zip(reversed(delimiters), reversed(sequences))

    # Format a single sequence of simple selectors, without combinator.
    def _sequence_str_(self) -> str:
        s = ""
        if self.tag:
            s += self.tag
        if self.classes:
            s += "".join(".%s" % class_ for class_ in self.classes)
        if self.id:
            s += "#%s" % self.id
        if self.attrs:
            s += "".join(str(attr) for attr in self.attrs)
        return s if s else "*"

    def from_str(cls, s: str, cursor: int = 0) -> Tuple["Selector", int]:
        Parses input string into selector.

        This factory function only parses out one selector (up to a
        comma or EOS), so partial consumption is allowed --- an optional
        `cursor` is taken as input (0 by default) and the moved cursor
        (either after the comma or at EOS) is returned as part of the

        :class:`SelectorParserException` is raised on invalid input. See
        :class:`Selector` documentation for the scope of support.

        If you need to completely consume a string representing
        (potentially) a group of selectors, use

            s:      input string
            cursor: initial cursor position on `s`

            A tuple containing the parsed selector and the moved the
            cursor (either after a comma-delimiter, or at EOS).
        # Simple selectors.
        TYPE_SEL = re.compile(r"[\w-]+", re.A)
        UNIVERSAL_SEL = re.compile(r"\*")
        ATTR_SEL = re.compile(
            re.A | re.X,
        CLASS_SEL = re.compile(r"\.([\w-]+)", re.A)
        ID_SEL = re.compile(r"#([\w-]+)", re.A)
        PSEUDO_CLASS_SEL = re.compile(r":[\w-]+(\([^)]+\))?", re.A)
        PSEUDO_ELEM_SEL = re.compile(r"::[\w-]+", re.A)

        # Combinators
        DESCENDANT_COM = re.compile(r"\s+")
        CHILD_COM = re.compile(r"\s*>\s*")
        NEXT_SIB_COM = re.compile(r"\s*\+\s*")
        SUB_SIB_COM = re.compile(r"\s*~\s*")

        # Misc
        WHITESPACE = re.compile(r"\s*")
        END_OF_SELECTOR = re.compile(r"\s*($|,)")

        tag = None
        classes = []
        id = None
        attrs = []
        combinator = None

        selector = None
        previous_combinator = None

        i = cursor

        # Skip leading whitespace
        m = WHITESPACE.match(s, i)
        if m:
            i = m.end()

        while i < len(s):
            # Parse one simple selector.
            # PEP 572 (assignment expressions; the one that burned Guido
            # so much that he resigned as BDFL) would have been nice; it
            # would have saved us from all the regex match
            # reassignments, and worse still, the casts, since mypy
            # complains about getting Optional[Match[str]] instead of
            # Match[str].
            if TYPE_SEL.match(s, i):
                if tag:
                    raise SelectorParserException(s, i, "multiple type selectors found")
                m = cast(Match[str], TYPE_SEL.match(s, i))
                tag = m.group()
            elif UNIVERSAL_SEL.match(s, i):
                m = cast(Match[str], UNIVERSAL_SEL.match(s, i))
            elif ATTR_SEL.match(s, i):
                m = cast(Match[str], ATTR_SEL.match(s, i))

                attr = m.group("attr")
                op = m.group("op")
                val_identifier = m.group("val_identifier")
                quote = m.group("quote")
                val_string_inner = m.group("val_string_inner")
                if val_identifier is not None:
                    val = val_identifier
                elif val_string_inner is not None:
                    val = val_string_inner.replace("\\" + quote, quote)
                    val = None

                if op is None:
                    type = AttributeSelectorType.BARE
                elif op == "=":
                    type = AttributeSelectorType.EQUAL
                elif op == "~=":
                    type = AttributeSelectorType.TILDE
                elif op == "|=":
                    type = AttributeSelectorType.PIPE
                elif op == "^=":
                    type = AttributeSelectorType.CARET
                elif op == "$=":
                    type = AttributeSelectorType.DOLLAR
                elif op == "*=":
                    type = AttributeSelectorType.ASTERISK
                else:  # pragma: no cover
                    raise SelectorParserException(
                        "unrecognized operator %s in attribute selector" % repr(op),

                attrs.append(AttributeSelector(attr, val, type))
            elif CLASS_SEL.match(s, i):
                m = cast(Match[str], CLASS_SEL.match(s, i))
            elif ID_SEL.match(s, i):
                if id:
                    raise SelectorParserException(s, i, "multiple id selectors found")
                m = cast(Match[str], ID_SEL.match(s, i))
                id = m.group(1)
            elif PSEUDO_CLASS_SEL.match(s, i):
                raise SelectorParserException(s, i, "pseudo-classes not supported")
            elif PSEUDO_ELEM_SEL.match(s, i):
                raise SelectorParserException(s, i, "pseudo-elements not supported")
                raise SelectorParserException(
                    s, i, "expecting simple selector, found none"
            i = m.end()

            # Try to parse a combinator, or end the selector.
            if CHILD_COM.match(s, i):
                m = cast(Match[str], CHILD_COM.match(s, i))
                combinator = Combinator.CHILD
            elif NEXT_SIB_COM.match(s, i):
                m = cast(Match[str], NEXT_SIB_COM.match(s, i))
                combinator = Combinator.NEXT_SIBLING
            elif SUB_SIB_COM.match(s, i):
                m = cast(Match[str], SUB_SIB_COM.match(s, i))
                combinator = Combinator.SUBSEQUENT_SIBLING
            elif END_OF_SELECTOR.match(s, i):
                m = cast(Match[str], END_OF_SELECTOR.match(s, i))
                combinator = None
            # Need to parse descendant combinator at the very end
            # because it could be a prefix to all previous cases.
            elif DESCENDANT_COM.match(s, i):
                m = cast(Match[str], DESCENDANT_COM.match(s, i))
                combinator = Combinator.DESCENDANT
            i = m.end()

            if combinator and i == len(s):
                raise SelectorParserException(s, i, "unexpected end at combinator")

            selector = cls(
            previous_combinator = combinator

            # End of selector.
            if combinator is None:

            tag = None
            classes = []
            id = None
            attrs = []
            combinator = None

        if not selector:
            raise SelectorParserException(s, i, "selector is empty")

        return selector, i

    def matches(self, node: "Node", root: Optional["Node"] = None) -> bool:
        Decides whether the selector matches `node`.

        Each sequence of simple selectors in the selector's chain must
        be matched for a positive.

        If `root` is provided and child and/or descendant combinators
        are involved, parent/ancestor lookup terminates at `root`.
        if self.tag:
            if not node.tag or node.tag != self.tag:
                return False
        if self.id:
            if node.attrs.get("id") != self.id:
                return False
        if self.classes:
            classes = node.classes
            for class_ in self.classes:
                if class_ not in classes:
                    return False
        if self.attrs:
            for attr_selector in self.attrs:
                if not attr_selector.matches(node):
                    return False

        if not self.previous:
            return True

        if self.combinator == Combinator.DESCENDANT:
            return any(
                self.previous.matches(ancestor, root=root)
                for ancestor in node.ancestors()
        elif self.combinator == Combinator.CHILD:
            if node is root or node.parent is None:
                return False
                return self.previous.matches(node.parent)
        elif self.combinator == Combinator.NEXT_SIBLING:
            sibling = node.previous_element_sibling()
            if not sibling:
                return False
                return self.previous.matches(sibling)
        elif self.combinator == Combinator.SUBSEQUENT_SIBLING:
            return any(
                self.previous.matches(sibling, root=root)
                for sibling in node.previous_siblings()
                if isinstance(sibling, ElementNode)
        else:  # pragma: no cover
            raise RuntimeError("unimplemented combinator: %s" % repr(self.combinator))

class AttributeSelector:
    Represents an attribute selector.

        attr (:class:`str`)
        val  (:class:`Optional`\\[:class:`str`])
        type (:class:`AttributeSelectorType`)

    def __init__(
        self, attr: str, val: Optional[str], type: "AttributeSelectorType"
    ) -> None:
        self.attr = attr.lower()
        self.val = val
        self.type = type

    def __repr__(self) -> str:
        return "<AttributeSelector %s>" % repr(str(self))

    def __str__(self) -> str:
        if self.type == AttributeSelectorType.BARE:
            fmt = "[{attr}{val:.0}]"
        elif self.type == AttributeSelectorType.EQUAL:
            fmt = "[{attr}={val}]"
        elif self.type == AttributeSelectorType.TILDE:
            fmt = "[{attr}~={val}]"
        elif self.type == AttributeSelectorType.PIPE:
            fmt = "[{attr}|={val}]"
        elif self.type == AttributeSelectorType.CARET:
            fmt = "[{attr}^={val}]"
        elif self.type == AttributeSelectorType.DOLLAR:
            fmt = "[{attr}$={val}]"
        elif self.type == AttributeSelectorType.ASTERISK:
            fmt = "[{attr}*={val}]"
        return fmt.format(attr=self.attr, val=repr(self.val))

    def matches(self, node: "Node") -> bool:
        val = node.attrs.get(self.attr)
        if val is None:
            return False
        if self.type == AttributeSelectorType.BARE:
            return True
        elif self.type == AttributeSelectorType.EQUAL:
            return val == self.val
        elif self.type == AttributeSelectorType.TILDE:
            return self.val in val.split()
        elif self.type == AttributeSelectorType.PIPE:
            return val == self.val or val.startswith("%s-" % self.val)
        elif self.type == AttributeSelectorType.CARET:
            return bool(self.val and val.startswith(self.val))
        elif self.type == AttributeSelectorType.DOLLAR:
            return bool(self.val and val.endswith(self.val))
        elif self.type == AttributeSelectorType.ASTERISK:
            return bool(self.val and self.val in val)
        else:  # pragma: no cover
            raise RuntimeError("unimplemented attribute selector: %s" % repr(self.type))

# Enum: basis for poor man's algebraic data type.
class AttributeSelectorType(Enum):
    Attribute selector types.

    Members correspond to the following forms of attribute selector:

    - :attr:`BARE`: ``[attr]``;
    - :attr:`EQUAL`: ``[attr=val]``;
    - :attr:`TILDE`: ``[attr~=val]``;
    - :attr:`PIPE`: ``[attr|=val]``;
    - :attr:`CARET`: ``[attr^=val]``;
    - :attr:`DOLLAR`: ``[attr$=val]``;
    - :attr:`ASTERISK`: ``[attr*=val]``.

    # [attr]
    BARE = 1
    # [attr=val]
    EQUAL = 2
    # [attr~=val]
    TILDE = 3
    # [attr|=val]
    PIPE = 4
    # [attr^=val]
    CARET = 5
    # [attr$=val]
    DOLLAR = 6
    # [attr*=val]
    ASTERISK = 7

class Combinator(Enum):
    Combinator types.

    Members correspond to the following combinators:

    - :attr:`DESCENDANT`: ``A B``;
    - :attr:`CHILD`: ``A > B``;
    - :attr:`NEXT_SIBLING`: ``A + B``;
    - :attr:`SUBSEQUENT_SIBLING`: ``A ~ B``.

    # ' '
    # >
    CHILD = 2
    # +
    # ~

def _tag_is_void(tag: str) -> bool:
    Checks whether the tag corresponds to a void element.

    return tag.lower() in (

def _tag_encloses_foreign_namespace(tag: str) -> bool:
    Checks whether the tag encloses a foreign namespace (MathML or SVG).

    return tag.lower() in ("math", "svg")

### end dim ###

# Global helper functions

def open_url(url):
    """Open an URL in the user's default web browser.

    The string attribute ``open_url.url_handler`` can be used to open URLs
    in a custom CLI script or utility. A subprocess is spawned with url as
    the parameter in this case instead of the usual webbrowser.open() call.

    Whether the browser's output (both stdout and stderr) are suppressed
    depends on the boolean attribute ``open_url.suppress_browser_output``.
    If the attribute is not set upon a call, set it to a default value,
    which means False if BROWSER is set to a known text-based browser --
    elinks, links, lynx, w3m or 'www-browser'; or True otherwise.

    The string attribute ``open_url.override_text_browser`` can be used to
    ignore env var BROWSER as well as some known text-based browsers and
    attempt to open url in a GUI browser available.
    Note: If a GUI browser is indeed found, this option ignores the program
          option `show-browser-logs`
    logger.debug('Opening %s', url)

    # Custom URL handler gets max priority
    if hasattr(open_url, 'url_handler'):
        subprocess.run([open_url.url_handler, url])

    browser = webbrowser.get()
    if open_url.override_text_browser:
        browser_output = open_url.suppress_browser_output
        for name in [b for b in webbrowser._tryorder if b not in text_browsers]:
            browser = webbrowser.get(name)

            # Found a GUI browser, suppress browser output
            open_url.suppress_browser_output = True

    if open_url.suppress_browser_output:
        _stderr = os.dup(2)
        _stdout = os.dup(1)
        # Patch for GUI browsers on WSL
        if "microsoft" not in platform.uname()[3].lower():
        fd = os.open(os.devnull, os.O_RDWR)
        os.dup2(fd, 2)
        os.dup2(fd, 1)
        browser.open(url, new=2)
        if open_url.suppress_browser_output:
            os.dup2(_stderr, 2)
            os.dup2(_stdout, 1)

    if open_url.override_text_browser:
        open_url.suppress_browser_output = browser_output

def printerr(msg):
    """Print message, verbatim, to stderr.

    ``msg`` could be any stringifiable value.
    print(msg, file=sys.stderr)

def unwrap(text):
    """Unwrap text."""
    lines = text.split('\n')
    result = ''
    for i in range(len(lines) - 1):
        result += lines[i]
        if not lines[i]:
            # Paragraph break
            result += '\n\n'
        elif lines[i + 1]:
            # Next line is not paragraph break, add space
            result += ' '
    # Handle last line
    result += lines[-1] if lines[-1] else '\n'
    return result

def check_stdout_encoding():
    """Make sure stdout encoding is utf-8.

    If not, print error message and instructions, then exit with
    status 1.

    This function is a no-op on win32 because encoding on win32 is
    messy, and let's just hope for the best. /s
    if sys.platform == 'win32':

    # Use codecs.lookup to resolve text encoding alias
    encoding = codecs.lookup(sys.stdout.encoding).name
    if encoding != 'utf-8':
        locale_lang, locale_encoding = locale.getlocale()
        if locale_lang is None:
            locale_lang = '<unknown>'
        if locale_encoding is None:
            locale_encoding = '<unknown>'
        ioencoding = os.getenv('PYTHONIOENCODING', 'not set')
        stdout encoding '{encoding}' detected. googler requires utf-8 to
        work properly. The wrong encoding may be due to a non-UTF-8
        locale or an improper PYTHONIOENCODING. (For the record, your
        locale language is {locale_lang} and locale encoding is
        {locale_encoding}; your PYTHONIOENCODING is {ioencoding}.)

        Please set a UTF-8 locale (e.g., en_US.UTF-8) or set
        PYTHONIOENCODING to utf-8.

def time_it(description=None):
    def decorator(func):
        def wrapped(*args, **kwargs):
            # Only profile in debug mode.
            if not logger.isEnabledFor(logging.DEBUG):
                return func(*args, **kwargs)

            import time
            mark = time.perf_counter()
            ret = func(*args, **kwargs)
            duration = time.perf_counter() - mark
            logger.debug('%s completed in \x1b[33m%.3fs\x1b[0m', description or func.__name__, duration)
            return ret

        return wrapped

    return decorator

# Classes

class HardenedHTTPSConnection(HTTPSConnection):
    """Overrides HTTPSConnection.connect to specify TLS version

    NOTE: TLS 1.2 is supported from Python 3.4

    def __init__(self, host, address_family=0, **kwargs):
        HTTPSConnection.__init__(self, host, **kwargs)
        self.address_family = address_family

    def connect(self, notweak=False):
        sock = self.create_socket_connection()

        # Optimizations not available on OS X
        if not notweak and sys.platform.startswith('linux'):
                sock.setsockopt(socket.SOL_TCP, socket.TCP_DEFER_ACCEPT, 1)
                sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_QUICKACK, 1)
                sock.setsockopt(socket.SOL_SOCKET, socket.SO_RCVBUF, 524288)
            except OSError:
                # Doesn't work on Windows' Linux subsystem (#179)
                logger.debug('setsockopt failed')

        if getattr(self, '_tunnel_host', None):
            self.sock = sock
        elif not notweak:
            # Try to use TLS 1.2
            ssl_context = None
            if hasattr(ssl, 'PROTOCOL_TLS'):
                # Since Python 3.5.3
                ssl_context = ssl.SSLContext(ssl.PROTOCOL_TLS)
                if hasattr(ssl_context, "minimum_version"):
                    # Python 3.7 with OpenSSL 1.1.0g or later
                    ssl_context.minimum_version = ssl.TLSVersion.TLSv1_2
                    ssl_context.options |= (ssl.OP_NO_SSLv2 | ssl.OP_NO_SSLv3 |
                                            ssl.OP_NO_TLSv1 | ssl.OP_NO_TLSv1_1)
            elif hasattr(ssl, 'PROTOCOL_TLSv1_2'):
                # Since Python 3.4
                ssl_context = ssl.SSLContext(ssl.PROTOCOL_TLSv1_2)
            if ssl_context:
                self.sock = ssl_context.wrap_socket(sock)

        # Fallback

    # Adapted from socket.create_connection.
    # https://github.com/python/cpython/blob/bce4ddafdd188cc6deb1584728b67b9e149ca6a4/Lib/socket.py#L771-L813
    def create_socket_connection(self):
        err = None
        results = socket.getaddrinfo(self.host, self.port, self.address_family, socket.SOCK_STREAM)
        # Prefer IPv4 if address family isn't explicitly specified.
        if self.address_family == 0:
            results = sorted(results, key=lambda res: 1 if res[0] == socket.AF_INET else 2)
        for af, socktype, proto, canonname, sa in results:
            sock = None
                sock = socket.socket(af, socktype, proto)
                if self.timeout is not None:
                if self.source_address:
                # Break explicitly a reference cycle
                err = None
                self.address_family = af
                logger.debug('Opened socket to %s:%d',
                             sa[0] if af == socket.AF_INET else ('[%s]' % sa[0]),
                return sock

            except socket.error as _:
                err = _
                if sock is not None:

        if err is not None:
                raise err
                # Break explicitly a reference cycle
                err = None
            raise socket.error("getaddrinfo returns an empty list")

class GoogleUrl(object):
    This class constructs the Google Search/News URL.

    This class is modelled on urllib.parse.ParseResult for familiarity,
    which means it supports reading of all six attributes -- scheme,
    netloc, path, params, query, fragment -- of
    urllib.parse.ParseResult, as well as the geturl() method.

    However, the attributes (properties) and methods listed below should
    be the preferred methods of access to this class.

    opts : dict or argparse.Namespace, optional
        See the ``opts`` parameter of `update`.

    Other Parameters
    See "Other Parameters" of `update`.

    hostname : str
        Read-write property.
    keywords : str or list of strs
        Read-write property.
    news : bool
        Read-only property.
    videos : bool
        Read-only property.
    url : str
        Read-only property.

    update(opts=None, **kwargs)


    def __init__(self, opts=None, **kwargs):
        self.scheme = 'https'
        # self.netloc is a calculated property
        self.path = '/search'
        self.params = ''
        # self.query is a calculated property
        self.fragment = ''

        self._tld = None
        self._num = 10
        self._start = 0
        self._keywords = []
        self._sites = None
        self._exclude = None

        self._query_dict = {
            'ie': 'UTF-8',
            'oe': 'UTF-8',
            #'gbv': '1',  # control the presence of javascript on the page, 1=no js, 2=js
            'sei': base64.encodebytes(uuid.uuid4().bytes).decode("ascii").rstrip('=\n').replace('/', '_'),

        # In preloaded HTML parsing mode, set keywords to something so
        # that we are not tripped up by require_keywords.
        if opts.html_file and not opts.keywords:
            opts.keywords = ['<debug>']

        self.update(opts, **kwargs)

    def __str__(self):
        return self.url

    def url(self):
        """The full Google URL you want."""
        return self.full()

    def hostname(self):
        """The hostname."""
        return self.netloc

    def hostname(self, hostname):
        self.netloc = hostname

    def keywords(self):
        """The keywords, either a str or a list of strs."""
        return self._keywords

    def keywords(self, keywords):
        self._keywords = keywords

    def news(self):
        """Whether the URL is for Google News."""
        return 'tbm' in self._query_dict and self._query_dict['tbm'] == 'nws'

    def videos(self):
        """Whether the URL is for Google Videos."""
        return 'tbm' in self._query_dict and self._query_dict['tbm'] == 'vid'

    def full(self):
        """Return the full URL.


        url = (self.scheme + ':') if self.scheme else ''
        url += '//' + self.netloc + self.relative()
        return url

    def relative(self):
        """Return the relative URL (without scheme and authority).

        Authority (see RFC 3986 section 3.2), or netloc in the
        terminology of urllib.parse, basically means the hostname
        here. The relative URL is good for making HTTP(S) requests to a
        known host.


        rel = self.path
        if self.params:
            rel += ';' + self.params
        if self.query:
            rel += '?' + self.query
        if self.fragment:
            rel += '#' + self.fragment
        return rel

    def update(self, opts=None, **kwargs):
        """Update the URL with the given options.

        opts : dict or argparse.Namespace, optional
            Carries options that affect the Google Search/News URL. The
            list of currently recognized option keys with expected value

                duration: str (GooglerArgumentParser.is_duration)
                exact: bool
                keywords: str or list of strs
                lang: str
                news: bool
                videos: bool
                num: int
                site: str
                start: int
                tld: str
                unfilter: bool

        Other Parameters
            The `kwargs` dict extends `opts`, that is, options can be
            specified either way, in `opts` or as individual keyword


        if opts is None:
            opts = {}
        if hasattr(opts, '__dict__'):
            opts = opts.__dict__

        qd = self._query_dict
        if opts.get('duration'):
            qd['tbs'] = 'qdr:%s' % opts['duration']
        if 'exact' in opts:
            if opts['exact']:
                qd['nfpr'] = 1
                qd.pop('nfpr', None)
        if opts.get('from') or opts.get('to'):
            cd_min = opts.get('from') or ''
            cd_max = opts.get('to') or ''
            qd['tbs'] = 'cdr:1,cd_min:%s,cd_max:%s' % (cd_min, cd_max)
        if 'keywords' in opts:
            self._keywords = opts['keywords']
        if 'lang' in opts and opts['lang']:
            qd['hl'] = opts['lang']
        if 'geoloc' in opts and opts['geoloc']:
            qd['gl'] = opts['geoloc']
        if 'news' in opts and opts['news']:
            qd['tbm'] = 'nws'
        elif 'videos' in opts and opts['videos']:
            qd['tbm'] = 'vid'
            qd.pop('tbm', None)
        if 'num' in opts:
            self._num = opts['num']
        if 'sites' in opts:
            self._sites = opts['sites']
        if 'exclude' in opts:
            self._exclude = opts['exclude']
        if 'start' in opts:
            self._start = opts['start']
        if 'tld' in opts:
            self._tld = opts['tld']
        if 'unfilter' in opts and opts['unfilter']:
            qd['filter'] = 0

    def set_queries(self, **kwargs):
        """Forcefully set queries outside the normal `update` mechanism.

        Other Parameters
            Arbitrary key value pairs to be set in the query string. All
            keys and values should be stringifiable.

            Note that certain keys, e.g., ``q``, have their values
            constructed on the fly, so setting those has no actual

        for k, v in kwargs.items():
            self._query_dict[k] = v

    def unset_queries(self, *args):
        """Forcefully unset queries outside the normal `update` mechanism.

        Other Parameters
            Arbitrary keys to be unset. No exception is raised if a key
            does not exist in the first place.

            Note that certain keys, e.g., ``q``, are always included in
            the resulting URL, so unsetting those has no actual effect.

        for k in args:
            self._query_dict.pop(k, None)

    def next_page(self):
        """Navigate to the next page."""
        self._start += self._num

    def prev_page(self):
        """Navigate to the previous page.

            If already at the first page (``start=0`` in the current
            query string).

        if self._start == 0:
            raise ValueError('Already at the first page.')
        self._start = (self._start - self._num) if self._start > self._num else 0

    def first_page(self):
        """Navigate to the first page.

            If already at the first page (``start=0`` in the current
            query string).

        if self._start == 0:
            raise ValueError('Already at the first page.')
        self._start = 0

    # Data source: https://web.archive.org/web/20170615200243/https://en.wikipedia.org/wiki/List_of_Google_domains
    # Scraper script: https://gist.github.com/zmwangx/b976e83c14552fe18b71
        'ac': 'google.ac',      'ad': 'google.ad',      'ae': 'google.ae',
        'af': 'google.com.af',  'ag': 'google.com.ag',  'ai': 'google.com.ai',
        'al': 'google.al',      'am': 'google.am',      'ao': 'google.co.ao',
        'ar': 'google.com.ar',  'as': 'google.as',      'at': 'google.at',
        'au': 'google.com.au',  'az': 'google.az',      'ba': 'google.ba',
        'bd': 'google.com.bd',  'be': 'google.be',      'bf': 'google.bf',
        'bg': 'google.bg',      'bh': 'google.com.bh',  'bi': 'google.bi',
        'bj': 'google.bj',      'bn': 'google.com.bn',  'bo': 'google.com.bo',
        'br': 'google.com.br',  'bs': 'google.bs',      'bt': 'google.bt',
        'bw': 'google.co.bw',   'by': 'google.by',      'bz': 'google.com.bz',
        'ca': 'google.ca',      'cat': 'google.cat',    'cc': 'google.cc',
        'cd': 'google.cd',      'cf': 'google.cf',      'cg': 'google.cg',
        'ch': 'google.ch',      'ci': 'google.ci',      'ck': 'google.co.ck',
        'cl': 'google.cl',      'cm': 'google.cm',      'cn': 'google.cn',
        'co': 'google.com.co',  'cr': 'google.co.cr',   'cu': 'google.com.cu',
        'cv': 'google.cv',      'cy': 'google.com.cy',  'cz': 'google.cz',
        'de': 'google.de',      'dj': 'google.dj',      'dk': 'google.dk',
        'dm': 'google.dm',      'do': 'google.com.do',  'dz': 'google.dz',
        'ec': 'google.com.ec',  'ee': 'google.ee',      'eg': 'google.com.eg',
        'es': 'google.es',      'et': 'google.com.et',  'fi': 'google.fi',
        'fj': 'google.com.fj',  'fm': 'google.fm',      'fr': 'google.fr',
        'ga': 'google.ga',      'ge': 'google.ge',      'gf': 'google.gf',
        'gg': 'google.gg',      'gh': 'google.com.gh',  'gi': 'google.com.gi',
        'gl': 'google.gl',      'gm': 'google.gm',      'gp': 'google.gp',
        'gr': 'google.gr',      'gt': 'google.com.gt',  'gy': 'google.gy',
        'hk': 'google.com.hk',  'hn': 'google.hn',      'hr': 'google.hr',
        'ht': 'google.ht',      'hu': 'google.hu',      'id': 'google.co.id',
        'ie': 'google.ie',      'il': 'google.co.il',   'im': 'google.im',
        'in': 'google.co.in',   'io': 'google.io',      'iq': 'google.iq',
        'is': 'google.is',      'it': 'google.it',      'je': 'google.je',
        'jm': 'google.com.jm',  'jo': 'google.jo',      'jp': 'google.co.jp',
        'ke': 'google.co.ke',   'kg': 'google.kg',      'kh': 'google.com.kh',
        'ki': 'google.ki',      'kr': 'google.co.kr',   'kw': 'google.com.kw',
        'kz': 'google.kz',      'la': 'google.la',      'lb': 'google.com.lb',
        'lc': 'google.com.lc',  'li': 'google.li',      'lk': 'google.lk',
        'ls': 'google.co.ls',   'lt': 'google.lt',      'lu': 'google.lu',
        'lv': 'google.lv',      'ly': 'google.com.ly',  'ma': 'google.co.ma',
        'md': 'google.md',      'me': 'google.me',      'mg': 'google.mg',
        'mk': 'google.mk',      'ml': 'google.ml',      'mm': 'google.com.mm',
        'mn': 'google.mn',      'ms': 'google.ms',      'mt': 'google.com.mt',
        'mu': 'google.mu',      'mv': 'google.mv',      'mw': 'google.mw',
        'mx': 'google.com.mx',  'my': 'google.com.my',  'mz': 'google.co.mz',
        'na': 'google.com.na',  'ne': 'google.ne',      'nf': 'google.com.nf',
        'ng': 'google.com.ng',  'ni': 'google.com.ni',  'nl': 'google.nl',
        'no': 'google.no',      'np': 'google.com.np',  'nr': 'google.nr',
        'nu': 'google.nu',      'nz': 'google.co.nz',   'om': 'google.com.om',
        'pa': 'google.com.pa',  'pe': 'google.com.pe',  'pg': 'google.com.pg',
        'ph': 'google.com.ph',  'pk': 'google.com.pk',  'pl': 'google.pl',
        'pn': 'google.co.pn',   'pr': 'google.com.pr',  'ps': 'google.ps',
        'pt': 'google.pt',      'py': 'google.com.py',  'qa': 'google.com.qa',
        'ro': 'google.ro',      'rs': 'google.rs',      'ru': 'google.ru',
        'rw': 'google.rw',      'sa': 'google.com.sa',  'sb': 'google.com.sb',
        'sc': 'google.sc',      'se': 'google.se',      'sg': 'google.com.sg',
        'sh': 'google.sh',      'si': 'google.si',      'sk': 'google.sk',
        'sl': 'google.com.sl',  'sm': 'google.sm',      'sn': 'google.sn',
        'so': 'google.so',      'sr': 'google.sr',      'st': 'google.st',
        'sv': 'google.com.sv',  'td': 'google.td',      'tg': 'google.tg',
        'th': 'google.co.th',   'tj': 'google.com.tj',  'tk': 'google.tk',
        'tl': 'google.tl',      'tm': 'google.tm',      'tn': 'google.tn',
        'to': 'google.to',      'tr': 'google.com.tr',  'tt': 'google.tt',
        'tw': 'google.com.tw',  'tz': 'google.co.tz',   'ua': 'google.com.ua',
        'ug': 'google.co.ug',   'uk': 'google.co.uk',   'uy': 'google.com.uy',
        'uz': 'google.co.uz',   'vc': 'google.com.vc',  've': 'google.co.ve',
        'vg': 'google.vg',      'vi': 'google.co.vi',   'vn': 'google.com.vn',
        'vu': 'google.vu',      'ws': 'google.ws',      'za': 'google.co.za',
        'zm': 'google.co.zm',   'zw': 'google.co.zw',

    def netloc(self):
        """The hostname."""
            return 'www.' + self.TLD_TO_DOMAIN_MAP[self._tld]
        except KeyError:
            return 'www.google.com'

    def query(self):
        """The query string."""
        qd = {}
        if self._num != 10:  # Skip sending the default
            qd['num'] = self._num
        if self._start:  # Skip sending the default
            qd['start'] = self._start

        # Construct the q query
        q = ''
        keywords = self._keywords
        sites = self._sites
        exclude = self._exclude
        if keywords:
            if isinstance(keywords, list):
                q += '+'.join(urllib.parse.quote_plus(kw) for kw in keywords)
                q += urllib.parse.quote_plus(keywords)
        if sites:
            q += '+OR'.join('+site:' + urllib.parse.quote_plus(site) for site in sites)
        if exclude:
            q += ''.join('+-site:' + urllib.parse.quote_plus(e) for e in exclude)
        qd['q'] = q
        return '&'.join('%s=%s' % (k, qd[k]) for k in sorted(qd.keys()))

class GoogleConnectionError(Exception):

class GoogleConnection(object):
    This class facilitates connecting to and fetching from Google.

    See http.client.HTTPSConnection for documentation of the


    host : str
        The currently connected host. Read-only property. Use
        `new_connection` to change host.

    new_connection(host=None, port=None, timeout=45)


    def __init__(self, host, port=None, address_family=0, timeout=45, proxy=None, notweak=False):
        self._host = None
        self._port = None
        self._address_family = address_family
        self._proxy = proxy
        self._notweak = notweak
        self._conn = None
        self.new_connection(host, port=port, timeout=timeout)
        self.cookie = ''

    def host(self):
        """The host currently connected to."""
        return self._host

    def new_connection(self, host=None, port=None, timeout=45):
        """Close the current connection (if any) and establish a new one.

        See http.client.HTTPSConnection for documentation of the
        parameters. Renew the connection (i.e., reuse the current host
        and port) if host is None or empty.


        if self._conn:

        if not host:
            host = self._host
            port = self._port
        self._host = host
        self._port = port
        host_display = host + (':%d' % port if port else '')

        proxy = self._proxy

        if proxy:
            proxy_user_passwd, proxy_host_port = parse_proxy_spec(proxy)

            logger.debug('Connecting to proxy server %s', proxy_host_port)
            self._conn = HardenedHTTPSConnection(proxy_host_port,
                                                 address_family=self._address_family, timeout=timeout)

            logger.debug('Tunnelling to host %s' % host_display)
            connect_headers = {}
            if proxy_user_passwd:
                connect_headers['Proxy-Authorization'] = 'Basic %s' % base64.b64encode(
            self._conn.set_tunnel(host, port=port, headers=connect_headers)

            except Exception as e:
                msg = 'Failed to connect to proxy server %s: %s.' % (proxy, e)
                raise GoogleConnectionError(msg)
            logger.debug('Connecting to new host %s', host_display)
            self._conn = HardenedHTTPSConnection(host, port=port,
                                                 address_family=self._address_family, timeout=timeout)
            except Exception as e:
                msg = 'Failed to connect to %s: %s.' % (host_display, e)
                raise GoogleConnectionError(msg)

    def renew_connection(self, timeout=45):
        """Renew current connection.

        Equivalent to ``new_connection(timeout=timeout)``.


    def fetch_page(self, url):
        """Fetch a URL.

        Allows one reconnection and multiple redirections before failing
        and raising GoogleConnectionError.

        url : str
            The URL to fetch, relative to the host.

            When not getting HTTP 200 even after the allowed one
            reconnection and/or one redirection, or when Google is
            blocking query due to unusual activity.

            Response payload, gunzipped (if applicable) and decoded (in UTF-8).

        except (http.client.HTTPException, OSError) as e:
            logger.debug('Got exception: %s.', e)
            logger.debug('Attempting to reconnect...')
            except http.client.HTTPException as e:
                logger.debug('Got exception: %s.', e)
                raise GoogleConnectionError("Failed to get '%s'." % url)

        resp = self._resp
        redirect_counter = 0
        while resp.status != 200 and redirect_counter < 3:
            if resp.status in {301, 302, 303, 307, 308}:
                redirection_url = resp.getheader('location', '')
                if 'sorry/IndexRedirect?' in redirection_url or 'sorry/index?' in redirection_url:
                    msg = "Connection blocked due to unusual activity.\n"
                    if self._conn.address_family == socket.AF_INET6:
                        msg += textwrap.dedent("""\
                        You are connecting over IPv6 which is likely the problem. Try to make
                        sure the machine has a working IPv4 network interface configured.
                        See also the -4, --ipv4 option of googler.\n""")
                    msg += textwrap.dedent("""\
                    THIS IS NOT A BUG, please do NOT report it as a bug unless you have specific
                    information that may lead to the development of a workaround.
                    You IP address is temporarily or permanently blocked by Google and requires
                    reCAPTCHA-solving to use the service, which googler is not capable of.
                    Possible causes include issuing too many queries in a short time frame, or
                    operating from a shared / low reputation IP with a history of abuse.
                    Please do NOT use googler for automated scraping.""")
                    msg = " ".join(msg.splitlines())
                    raise GoogleConnectionError(msg)
                resp = self._resp
                redirect_counter += 1

        if resp.status != 200:
            raise GoogleConnectionError('Got HTTP %d: %s' % (resp.status, resp.reason))

        payload = resp.read()
            return gzip.decompress(payload).decode('utf-8')
        except OSError:
            # Not gzipped
            return payload.decode('utf-8')

    def _redirect(self, url):
        """Redirect to and fetch a new URL.

        Like `_raw_get`, the response is stored in ``self._resp``. A new
        connection is made if redirecting to a different host.

        url : str
            If absolute and points to a different host, make a new


        logger.debug('Redirecting to URL %s', url)
        segments = urllib.parse.urlparse(url)

        host = segments.netloc
        if host != self._host:

        relurl = urllib.parse.urlunparse(('', '') + segments[2:])
        except http.client.HTTPException as e:
            logger.debug('Got exception: %s.', e)
            raise GoogleConnectionError("Failed to get '%s'." % url)

    def _raw_get(self, url):
        """Make a raw HTTP GET request.

        No status check (which implies no redirection). Response can be
        accessed from ``self._resp``.

        url : str
            URL relative to the host, used in the GET request.


        logger.debug('Fetching URL %s', url)
        self._conn.request('GET', url, None, {
            'Accept': 'text/html',
            'Accept-Encoding': 'gzip',
            'User-Agent': USER_AGENT,
            'Cookie': self.cookie,
            'Connection': 'keep-alive',
            'DNT': '1',
        self._resp = self._conn.getresponse()
        if self.cookie == '':
            complete_cookie = self._resp.getheader('Set-Cookie')
            # Cookie won't be available if already blocked
            if complete_cookie is not None:
                self.cookie = complete_cookie[:complete_cookie.find(';')]
                logger.debug('Cookie: %s' % self.cookie)

    def close(self):
        """Close the connection (if one is active)."""
        if self._conn:

class GoogleParser(object):

    def __init__(self, html, *, news=False, videos=False):
        self.news = news
        self.videos = videos
        self.autocorrected = False
        self.showing_results_for = None
        self.filtered = False
        self.results = []

    def parse(self, html):
        tree = parse_html(html)

        if debugger:
            printerr('\x1b[1mInspect the DOM through the \x1b[4mtree\x1b[24m variable.\x1b[0m')
                import IPython
            except ImportError:
                import pdb

        # cw is short for collapse_whitespace.
        cw = lambda s: re.sub(r'[ \t\n\r]+', ' ', s) if s is not None else s

        index = 0
        for div_g in tree.select_all('div.g'):
            if div_g.select('.hp-xpdbox'):
                # Skip smart cards.
                if div_g.select('.st'):
                    # Old class structure, stopped working some time in
                    # September 2020, but kept just in case.
                    h3 = div_g.select('div.r h3')
                    if h3:
                        title = h3.text
                        a = h3.parent
                        h3 = div_g.select('h3.r')
                        a = h3.select('a')
                        title = a.text
                        mime = div_g.select('.mime')
                        if mime:
                            title = mime.text + ' ' + title
                    abstract_node = div_g.select('.st')
                    metadata_node = div_g.select('.f')
                    # Current structure as of October 2020.
                    # Note that a filetype tag (e.g. PDF) is now pretty
                    # damn hard to parse with confidence (that it'll
                    # survive the slighest further change), so we don't.

                    # As of January 15th 2021, the html class is not rc anymore, it's tF2Cxc.
                    # This approach is not very resilient to changes by Google, but it works for now.
                    # title_node, details_node, *_ = div_g.select_all('div.rc > div')
                    title_node, details_node, *_ = div_g.select_all('div.tF2Cxc > div')
                    if 'yuRUbf' not in title_node.classes:
                        logger.debug('unexpected title node class(es): expected %r, got %r',
                                     'yuRUbf', ' '.join(title_node.classes))
                    if 'IsZvec' not in details_node.classes:
                        logger.debug('unexpected details node class(es): expected %r, got %r',
                                     'IsZvec', ' '.join(details_node.classes))
                    a = title_node.select('a')
                    h3 = a.select('h3')
                    title = h3.text
                    abstract_node = details_node.select('span')
                    metadata_node = details_node.select('.f, span ~ div')
                url = self.unwrap_link(a.attr('href'))
                matched_keywords = []
                abstract = ''
                # BFS descendant nodes. Necessary to locate matches (b,
                # em) while skipping metadata (.f).
                abstract_nodes = collections.deque([abstract_node])
                while abstract_nodes:
                    node = abstract_nodes.popleft()
                    if 'f' in node.classes:
                        # .f is handled as metadata instead.
                    if node.tag in ['b', 'em']:
                        matched_keywords.append({'phrase': node.text, 'offset': len(abstract)})
                        abstract += node.text
                    if not node.children:
                        abstract += node.text
                    for child in node.children:
                metadata = None
                    # Sometimes there are multiple metadata fields
                    # associated with a single entry, e.g. "Released",
                    # "Producer(s)", "Genre", etc. for a song (sample
                    # query: "never gonna give you up"). These need to
                    # be delimited when displayed.
                    metadata_fields = metadata_node.select_all('div > div.wFMWsc')
                    if metadata_fields:
                        metadata = ' | '.join(field.text for field in metadata_fields)
                    elif not metadata_node.select('a') and not metadata_node.select('g-expandable-container'):
                        metadata = metadata_node.text
                    if metadata:
                        metadata = (
                            .replace('\u200e', '')
                            .replace(' - ', ', ')
                            .replace(' \u2014 ', ', ')
                except AttributeError:
            except (AttributeError, ValueError):
            sitelinks = []
            for td in div_g.select_all('td'):
                    a = td.select('a')
                    sl_title = a.text
                    sl_url = self.unwrap_link(a.attr('href'))
                    sl_abstract = td.select('div.s.st, div.s .st').text
                    sitelink = Sitelink(cw(sl_title), sl_url, cw(sl_abstract))
                    if sitelink not in sitelinks:
                except (AttributeError, ValueError):
            # cw cannot be applied to abstract here since it may screw
            # up offsets of matches. Instead, each relevant node's text
            # is whitespace-collapsed before being appended to abstract.
            # We then hope for the best.
            result = Result(index + 1, cw(title), url, abstract,
                            metadata=cw(metadata), sitelinks=sitelinks, matches=matched_keywords)
            if result not in self.results:
                index += 1

        if not self.results:
            for card in tree.select_all('g-card'):
                a = card.select('a[href]')
                if not a:
                url = self.unwrap_link(a.attr('href'))
                text_nodes = []
                for node in a.descendants():
                    if isinstance(node, TextNode) and node.strip():
                if len(text_nodes) != 4:
                publisher, title, abstract, publishing_time = text_nodes
                metadata = '%s, %s' % (publisher, publishing_time)
                index += 1
                self.results.append(Result(index, cw(title), url, cw(abstract), metadata=cw(metadata)))

        # Showing results for ...
        # Search instead for ...
        spell_orig = tree.select("span.spell_orig")
        if spell_orig:
            showing_results_for_link = next(
                filter(lambda el: el.tag == "a", spell_orig.previous_siblings()), None
            if showing_results_for_link:
                self.autocorrected = True
                self.showing_results_for = showing_results_for_link.text

        # No results found for ...
        # Results for ...:
        alt_query_infobox = tree.select('#topstuff')
        if alt_query_infobox:
            bolds = alt_query_infobox.select_all('div b')
            if len(bolds) == 2:
                self.showing_results_for = bolds[1].text

        # In order to show you the most relevant results, we have
        # omitted some entries very similar to the N already displayed.
        # ...
        self.filtered = tree.select('p#ofr') is not None

    # Unwraps /url?q=http://...&sa=...
    # TODO: don't unwrap if URL isn't in this form.
    def unwrap_link(link):
        qs = urllib.parse.urlparse(link).query
            url = urllib.parse.parse_qs(qs)['q'][0]
        except KeyError:
            return link
            if "://" in url:
                return url
                # Google's internal services link, e.g.,
                # /search?q=google&..., which cannot be unwrapped into
                # an actual URL.
                raise ValueError(link)

class Sitelink(object):
    """Container for a sitelink."""

    def __init__(self, title, url, abstract):
        self.title = title
        self.url = url
        self.abstract = abstract
        self.index = ''

    def __eq__(self, other):
        return (
            self.title == other.title and
            self.url == other.url and
            self.abstract == other.abstract

    def __hash__(self):
        return hash((self.title, self.url, self.abstract))

Colors = collections.namedtuple('Colors', 'index, title, url, metadata, abstract, prompt, reset')

class Result(object):
    Container for one search result, with output helpers.

    index : int or str
    title : str
    url : str
    abstract : str
    metadata : str, optional
        Only applicable to Google News results, with publisher name and
        publishing time.
    sitelinks : list, optional
        List of ``SiteLink`` objects.

    index : str
    title : str
    url : str
    abstract : str
    metadata : str or None
    sitelinks : list
    matches : list

    Class Variables
    colors : str



    # Class variables
    colors = None
    urlexpand = True

    def __init__(self, index, title, url, abstract, metadata=None, sitelinks=None, matches=None):
        index = str(index)
        self.index = index
        self.title = title
        self.url = url
        self.abstract = abstract
        self.metadata = metadata
        self.sitelinks = [] if sitelinks is None else sitelinks
        self.matches = [] if matches is None else matches

        self._urltable = {index: url}
        subindex = 'a'
        for sitelink in self.sitelinks:
            fullindex = index + subindex
            sitelink.index = fullindex
            self._urltable[fullindex] = sitelink.url
            subindex = chr(ord(subindex) + 1)

    def __eq__(self, other):
        return (
            self.title == other.title and
            self.url == other.url and
            self.abstract == other.abstract and
            self.metadata == other.metadata and
            self.sitelinks == other.sitelinks and
            self.matches == other.matches

    def __hash__(self):
        sitelinks_hashable = tuple(self.sitelinks) if self.sitelinks is not None else None
        matches_hashable = tuple(self.matches) if self.matches is not None else None
        return hash(self.title, self.url, self.abstract, self.metadata, sitelinks_hashable, matches_hashable)

    def _print_title_and_url(self, index, title, url, indent=0):
        colors = self.colors

        if not self.urlexpand:
            url = '[' + urllib.parse.urlparse(url).netloc + ']'

        if colors:
            # Adjust index to print result index clearly
            print(" %s%s%-3s%s" % (' ' * indent, colors.index, index + '.', colors.reset), end='')
            if not self.urlexpand:
                print(' ' + colors.title + title + colors.reset + ' ' + colors.url + url + colors.reset)
                print(' ' + colors.title + title + colors.reset)
                print(' ' * (indent + 5) + colors.url + url + colors.reset)
            if self.urlexpand:
                print(' %s%-3s %s' % (' ' * indent, index + '.', title))
                print(' %s%s' % (' ' * (indent + 4), url))
                print(' %s%-3s %s %s' % (' ' * indent, index + '.', title, url))

    def _print_metadata_and_abstract(self, abstract, metadata=None, matches=None, indent=0):
        colors = self.colors
            columns, _ = os.get_terminal_size()
        except OSError:
            columns = 0

        if metadata:
            if colors:
                print(' ' * (indent + 5) + colors.metadata + metadata + colors.reset)
                print(' ' * (indent + 5) + metadata)

        if abstract:
            fillwidth = (columns - (indent + 6)) if columns > indent + 6 else len(abstract)
            wrapped_abstract = TrackedTextwrap(abstract, fillwidth)
            if colors:
                # Highlight matches.
                for match in matches or []:
                    offset = match['offset']
                    span = len(match['phrase'])
                    wrapped_abstract.insert_zero_width_sequence('\x1b[1m', offset)
                    wrapped_abstract.insert_zero_width_sequence('\x1b[0m', offset + span)

            if colors:
                print(colors.abstract, end='')
            for line in wrapped_abstract.lines:
                print('%s%s' % (' ' * (indent + 5), line))
            if colors:
                print(colors.reset, end='')


    def print(self):
        """Print the result entry."""
        self._print_title_and_url(self.index, self.title, self.url)
        self._print_metadata_and_abstract(self.abstract, metadata=self.metadata, matches=self.matches)

        for sitelink in self.sitelinks:
            self._print_title_and_url(sitelink.index, sitelink.title, sitelink.url, indent=4)
            self._print_metadata_and_abstract(sitelink.abstract, indent=4)

    def jsonizable_object(self):
        """Return a JSON-serializable dict representing the result entry."""
        obj = {
            'title': self.title,
            'url': self.url,
            'abstract': self.abstract
        if self.metadata:
            obj['metadata'] = self.metadata
        if self.sitelinks:
            obj['sitelinks'] = [sitelink.__dict__ for sitelink in self.sitelinks]
        if self.matches:
            obj['matches'] = self.matches
        return obj

    def urltable(self):
        """Return a index-to-URL table for the current result.

        Normally, the table contains only a single entry, but when the result
        contains sitelinks, all sitelinks are included in this table.

            A dict mapping indices (strs) to URLs (also strs). Indices of
            sitelinks are the original index appended by lowercase letters a,
            b, c, etc.

        return self._urltable

    def collapse_whitespace(s):
        return re.sub(r'[ \t\n\r]+', ' ', s)

class GooglerCmdException(Exception):

class NoKeywordsException(GooglerCmdException):

def require_keywords(method):
    # Require keywords to be set before we run a GooglerCmd method. If
    # no keywords have been set, raise a NoKeywordsException.
    def enforced_method(self, *args, **kwargs):
        if not self.keywords:
            raise NoKeywordsException('No keywords.')
        method(self, *args, **kwargs)

    return enforced_method

def no_argument(method):
    # Normalize a do_* method of GooglerCmd that takes no argument to
    # one that takes an arg, but issue a warning when an nonempty
    # argument is given.
    def enforced_method(self, arg):
        if arg:
            method_name = arg.__name__
            command_name = method_name[3:] if method_name.startswith('do_') else method_name
            logger.warning("Argument to the '%s' command ignored.", command_name)

    return enforced_method

class GooglerCmd(object):
    Command line interpreter and executor class for googler.

    Inspired by PSL cmd.Cmd.

    opts : argparse.Namespace
        Options and/or arguments.

    options : argparse.Namespace
        Options that are currently in effect. Read-only attribute.
    keywords : str or list or strs
        Current keywords. Read-only attribute

    display_results(prelude='\n', json_output=False)
    fetch_and_display(prelude='\n', json_output=False, interactive=True)

    # Class variables
    colors = None
    re_url_index = re.compile(r"\d+(a-z)?")

    def __init__(self, opts):

        self._opts = opts

        self._google_url = GoogleUrl(opts)

        if opts.html_file:
            # Preloaded HTML parsing mode, do not initialize connection.
            self._preload_from_file = opts.html_file
            self._conn = None
            self._preload_from_file = None
            proxy = opts.proxy if hasattr(opts, 'proxy') else None
            self._conn = GoogleConnection(self._google_url.hostname,

        self.results = []
        self._autocorrected = None
        self._showing_results_for = None
        self._results_filtered = False
        self._urltable = {}

        self.promptcolor = True if os.getenv('DISABLE_PROMPT_COLOR') is None else False

        self.no_results_instructions_shown = False

    def options(self):
        """Current options."""
        return self._opts

    def keywords(self):
        """Current keywords."""
        return self._google_url.keywords

    def fetch(self):
        """Fetch a page and parse for results.

        Results are stored in ``self.results``.


        See Also

        # This method also sets self._results_filtered and
        # self._urltable.
        if self._preload_from_file:
            with open(self._preload_from_file, encoding='utf-8') as fp:
                page = fp.read()
            page = self._conn.fetch_page(self._google_url.relative())
            if logger.isEnabledFor(logging.DEBUG):
                import tempfile
                fd, tmpfile = tempfile.mkstemp(prefix='googler-response-', suffix='.html')
                with open(tmpfile, 'w', encoding='utf-8') as fp:
                logger.debug("Response body written to '%s'.", tmpfile)

        parser = GoogleParser(page, news=self._google_url.news, videos=self._google_url.videos)

        self.results = parser.results
        self._autocorrected = parser.autocorrected
        self._showing_results_for = parser.showing_results_for
        self._results_filtered = parser.filtered
        self._urltable = {}
        for r in self.results:

    def warn_no_results(self):
        printerr('No results.')
        if self.no_results_instructions_shown:

            import json
            import urllib.error
            import urllib.request
            info_json_url = '%s/master/info.json' % RAW_DOWNLOAD_REPO_BASE
            logger.debug('Fetching %s for project status...', info_json_url)
                with urllib.request.urlopen(info_json_url, timeout=5) as response:
                        info = json.load(response)
                    except Exception:
                        logger.error('Failed to decode project status from %s', info_json_url)
                        raise RuntimeError
            except urllib.error.HTTPError as e:
                logger.error('Failed to fetch project status from %s: HTTP %d', info_json_url, e.code)
                raise RuntimeError
            epoch = info.get('epoch')
            if epoch > _EPOCH_:
                printerr('Your version of googler is broken due to Google-side changes.')
                tracking_issue = info.get('tracking_issue')
                fixed_on_master = info.get('fixed_on_master')
                fixed_in_release = info.get('fixed_in_release')
                if fixed_in_release:
                    printerr('A new version, %s, has been released to address the changes.' % fixed_in_release)
                    printerr('Please upgrade to the latest version.')
                elif fixed_on_master:
                    printerr('The fix has been pushed to master, pending a release.')
                    printerr('Please download the master version https://git.io/googler or wait for a release.')
                    printerr('The issue is tracked at https://github.com/jarun/googler/issues/%s.' % tracking_issue)
        except RuntimeError:

        printerr('If you believe this is a bug, please review '
                 'https://git.io/googler-no-results before submitting a bug report.')
        self.no_results_instructions_shown = True

    def display_results(self, prelude='\n', json_output=False):
        """Display results stored in ``self.results``.

        See `fetch_and_display`.

        if json_output:
            # JSON output
            import json
            results_object = [r.jsonizable_object() for r in self.results]
            print(json.dumps(results_object, indent=2, sort_keys=True, ensure_ascii=False))
            # Regular output
            if not self.results:
                for r in self.results:

    def showing_results_for_alert(self, interactive=True):
        colors = self.colors
        if self._showing_results_for:
            if colors:
                # Underline the query
                actual_query = '\x1b[4m' + self._showing_results_for + '\x1b[24m'
                actual_query = self._showing_results_for
            if self._autocorrected:
                if interactive:
                    info = 'Showing results for %s; enter "x" for an exact search.' % actual_query
                    info = 'Showing results for %s; use -x, --exact for an exact search.' % actual_query
                info = 'No results found; showing results for %s.' % actual_query
            if interactive:
            if colors:
                printerr(colors.prompt + info + colors.reset)
                printerr('** ' + info)

    def fetch_and_display(self, prelude='\n', json_output=False, interactive=True):
        """Fetch a page and display results.

        Results are stored in ``self.results``.

        prelude : str, optional
            A string that is written to stderr before showing actual results,
            usually serving as a separator. Default is an empty line.
        json_output : bool, optional
            Whether to dump results in JSON format. Default is False.
        interactive : bool, optional
            Whether to show contextual instructions, when e.g. Google
            has filtered the results. Default is True.


        See Also

        self.display_results(prelude=prelude, json_output=json_output)
        if self._results_filtered:
            colors = self.colors
            info = 'Enter "unfilter" to show similar results Google omitted.'
            if colors:
                printerr(colors.prompt + info + colors.reset)
                printerr('** ' + info)

    def read_next_command(self):
        """Show omniprompt and read user command line.

        Command line is always stripped, and each consecutive group of
        whitespace is replaced with a single space character. If the
        command line is empty after stripping, when ignore it and keep
        reading. Exit with status 0 if we get EOF or an empty line
        (pre-strip, that is, a raw <enter>) twice in a row.

        The new command line (non-empty) is stored in ``self.cmd``.

        colors = self.colors
        message = 'googler (? for help)'
        prompt = (colors.prompt + message + colors.reset + ' ') if (colors and self.promptcolor) else (message + ': ')
        enter_count = 0
        while True:
                cmd = input(prompt)
            except EOFError:

            if not cmd:
                enter_count += 1
                if enter_count == 2:
                    # Double <enter>
                enter_count = 0

            cmd = ' '.join(cmd.split())
            if cmd:
                self.cmd = cmd

    def help():

    def do_first(self):
        except ValueError as e:
            print(e, file=sys.stderr)


    def do_google(self, arg):
        # Update keywords and reconstruct URL
        self._opts.keywords = arg
        self._google_url = GoogleUrl(self._opts)

    def do_next(self):
        # If > 5 results are being fetched each time,
        # block next when no parsed results in current fetch
        if not self.results and self._google_url._num > 5:
            printerr('No results.')

    def do_open(self, *args):
        if not args:

        for nav in args:
            if nav == 'a':
                for key, value in sorted(self._urltable.items()):
            elif nav in self._urltable:
            elif '-' in nav:
                    vals = [int(x) for x in nav.split('-')]
                    if (len(vals) != 2):
                        printerr('Invalid range %s.' % nav)

                    if vals[0] > vals[1]:
                        vals[0], vals[1] = vals[1], vals[0]

                    for _id in range(vals[0], vals[1] + 1):
                        if str(_id) in self._urltable:
                            printerr('Invalid index %s.' % _id)
                except ValueError:
                    printerr('Invalid range %s.' % nav)
                printerr('Invalid index %s.' % nav)

    def do_previous(self):
        except ValueError as e:
            print(e, file=sys.stderr)


    def do_exact(self):
        # Reset start to 0 when exact is applied.
        self._google_url.update(start=0, exact=True)

    def do_unfilter(self):
        # Reset start to 0 when unfilter is applied.

    def copy_url(self, idx):
                content = self._urltable[idx].encode('utf-8')
            except KeyError:
                printerr('Invalid index.')

            # try copying the url to clipboard using native utilities
            copier_params = []
            if sys.platform.startswith(('linux', 'freebsd', 'openbsd')):
                if shutil.which('xsel') is not None:
                    copier_params = ['xsel', '-b', '-i']
                elif shutil.which('xclip') is not None:
                    copier_params = ['xclip', '-selection', 'clipboard']
                elif shutil.which('wl-copy') is not None:
                    copier_params = ['wl-copy']
                elif shutil.which('termux-clipboard-set') is not None:
                    copier_params = ['termux-clipboard-set']
            elif sys.platform == 'darwin':
                copier_params = ['pbcopy']
            elif sys.platform == 'win32':
                copier_params = ['clip']

            if copier_params:
                Popen(copier_params, stdin=PIPE, stdout=DEVNULL, stderr=DEVNULL).communicate(content)

            # If native clipboard utilities are absent, try to use terminal multiplexers
            # tmux
            if os.getenv('TMUX_PANE'):
                copier_params = ['tmux', 'set-buffer']
                Popen(copier_params + [content], stdin=DEVNULL, stdout=DEVNULL, stderr=DEVNULL).communicate()

            # GNU Screen paste buffer
            if os.getenv('STY'):
                import tempfile
                copier_params = ['screen', '-X', 'readbuf', '-e', 'utf8']
                tmpfd, tmppath = tempfile.mkstemp()
                    with os.fdopen(tmpfd, 'wb') as fp:
                    Popen(copier_params, stdin=DEVNULL, stdout=DEVNULL, stderr=DEVNULL).communicate()

            printerr('failed to locate suitable clipboard utility')
        except Exception:
            raise NoKeywordsException

    def cmdloop(self):
        """Run REPL."""
        if self.keywords:
            printerr('Please initiate a query.')

        while True:
            # TODO: Automatic dispatcher
            # We can't write a dispatcher for now because that could
            # change behaviour of the prompt. However, we have already
            # laid a lot of ground work for the dispatcher, e.g., the
            # `no_argument' decorator.
                cmd = self.cmd
                if cmd == 'f':
                elif cmd.startswith('g '):
                elif cmd == 'n':
                elif cmd == 'o':
                elif cmd.startswith('o '):
                elif cmd.startswith('O '):
                    open_url.override_text_browser = True
                    open_url.override_text_browser = False
                elif cmd == 'p':
                elif cmd == 'q':
                elif cmd == 'x':
                elif cmd == 'unfilter':
                elif cmd == '?':
                elif cmd in self._urltable:
                elif self.keywords and cmd.isdigit() and int(cmd) < 100:
                    printerr('Index out of bound. To search for the number, use g.')
                elif cmd == 'u':
                    Result.urlexpand = not Result.urlexpand
                elif cmd.startswith('c ') and self.re_url_index.match(cmd[2:]):
            except NoKeywordsException:
                printerr('Initiate a query first.')

class GooglerArgumentParser(argparse.ArgumentParser):
    """Custom argument parser for googler."""

    # Print omniprompt help
    def print_omniprompt_help(file=None):
        file = sys.stderr if file is None else file
        omniprompt keys:
          n, p                  fetch the next or previous set of search results
          index                 open the result corresponding to index in browser
          f                     jump to the first page
          o [index|range|a ...] open space-separated result indices, numeric ranges
                                (sitelinks unsupported in ranges), or all, in browser
                                open the current search in browser, if no arguments
          O [index|range|a ...] like key 'o', but try to open in a GUI browser
          g keywords            new Google search for 'keywords' with original options
                                should be used to search omniprompt keys and indices
          c index               copy url to clipboard
          u                     toggle url expansion
          q, ^D, double Enter   exit googler
          ?                     show omniprompt help
          *                     other inputs issue a new search with original options

    # Print information on googler
    def print_general_info(file=None):
        file = sys.stderr if file is None else file
        Version %s
        Copyright © 2008 Henri Hakkinen
        Copyright © 2015-2021 Arun Prakash Jana <[email protected]>
        Zhiming Wang <[email protected]>
        License: GPLv3
        Webpage: https://github.com/jarun/googler
        """ % _VERSION_))

    # Augment print_help to print more than synopsis and options
    def print_help(self, file=None):

    # Automatically print full help text on error
    def error(self, message):
        sys.stderr.write('%s: error: %s\n\n' % (self.prog, message))

    # Type guards
    def positive_int(arg):
        """Try to convert a string into a positive integer."""
            n = int(arg)
            assert n > 0
            return n
        except (ValueError, AssertionError):
            raise argparse.ArgumentTypeError('%s is not a positive integer' % arg)

    def nonnegative_int(arg):
        """Try to convert a string into a nonnegative integer."""
            n = int(arg)
            assert n >= 0
            return n
        except (ValueError, AssertionError):
            raise argparse.ArgumentTypeError('%s is not a non-negative integer' % arg)

    def is_duration(arg):
        """Check if a string is a valid duration accepted by Google.

        A valid duration is of the form dNUM, where d is a single letter h
        (hour), d (day), w (week), m (month), or y (year), and NUM is a
        non-negative integer.
            if arg[0] not in ('h', 'd', 'w', 'm', 'y') or int(arg[1:]) < 0:
                raise ValueError
        except (TypeError, IndexError, ValueError):
            raise argparse.ArgumentTypeError('%s is not a valid duration' % arg)
        return arg

    def is_date(arg):
        """Check if a string is a valid date/month/year accepted by Google."""
        if re.match(r'^(\d+/){0,2}\d+$', arg):
            return arg
            raise argparse.ArgumentTypeError('%s is not a valid date/month/year; '
                                             'use the American date format with slashes')

    def is_colorstr(arg):
        """Check if a string is a valid color string."""
            assert len(arg) == 6
            for c in arg:
                assert c in COLORMAP
        except AssertionError:
            raise argparse.ArgumentTypeError('%s is not a valid color string' % arg)
        return arg

# Self-upgrade mechanism

def system_is_windows():
    """Checks if the underlying system is Windows (Cygwin included)."""
    return sys.platform in {'win32', 'cygwin'}

def get_latest_ref(include_git=False):
    """Helper for download_latest_googler."""
    import urllib.request

    if include_git:
        # Get SHA of latest commit on master
        request = urllib.request.Request('%s/commits/master' % API_REPO_BASE,
                                         headers={'Accept': 'application/vnd.github.v3.sha'})
        response = urllib.request.urlopen(request)
        if response.status != 200:
            raise http.client.HTTPException(response.reason)
        return response.read().decode('utf-8')
        # Get name of latest tag
        request = urllib.request.Request('%s/releases?per_page=1' % API_REPO_BASE,
                                         headers={'Accept': 'application/vnd.github.v3+json'})
        response = urllib.request.urlopen(request)
        if response.status != 200:
            raise http.client.HTTPException(response.reason)
        import json
        return json.loads(response.read().decode('utf-8'))[0]['tag_name']

def download_latest_googler(include_git=False):
    """Download latest googler to a temp file.

    By default, the latest released version is downloaded, but if
    `include_git` is specified, then the latest git master is downloaded

    include_git : bool, optional
        Download from git master. Default is False.

    (git_ref, path): tuple
         A tuple containing the git reference (either name of the latest
         tag or SHA of the latest commit) and path to the downloaded

    # Download googler to a tempfile
    git_ref = get_latest_ref(include_git=include_git)
    googler_download_url = '%s/%s/googler' % (RAW_DOWNLOAD_REPO_BASE, git_ref)
    printerr('Downloading %s' % googler_download_url)
    request = urllib.request.Request(googler_download_url,
                                     headers={'Accept-Encoding': 'gzip'})
    import tempfile
    fd, path = tempfile.mkstemp()
    atexit.register(lambda: os.remove(path) if os.path.exists(path) else None)
    with open(path, 'wb') as fp:
        with urllib.request.urlopen(request) as response:
            if response.status != 200:
                raise http.client.HTTPException(response.reason)
            payload = response.read()
            except OSError:
    return git_ref, path

def self_replace(path):
    """Replace the current script with a specified file.

    Both paths (the specified path and path to the current script) are
    resolved to absolute, symlink-free paths. Upon replacement, the
    owner and mode signatures of the current script are preserved. The
    caller needs to have the necessary permissions.

    Replacement won't happen if the specified file is the same
    (content-wise) as the current script.

    path : str
        Path to the replacement file.

        True if replaced, False if skipped (specified file is the same
        as the current script).

    if system_is_windows():
        raise NotImplementedError('Self upgrade not supported on Windows.')

    import filecmp
    import shutil

    path = os.path.realpath(path)
    self_path = os.path.realpath(__file__)

    if filecmp.cmp(path, self_path):
        return False

    self_stat = os.stat(self_path)
    os.chown(path, self_stat.st_uid, self_stat.st_gid)
    os.chmod(path, self_stat.st_mode)

    shutil.move(path, self_path)
    return True

def self_upgrade(include_git=False):
    """Perform in-place self-upgrade.

    include_git : bool, optional
        See `download_latest_googler`. Default is False.

    git_ref, path = download_latest_googler(include_git=include_git)
    if self_replace(path):
        printerr('Upgraded to %s.' % git_ref)
        printerr('Already up to date.')

def check_new_version():
        from distutils.version import StrictVersion as Version
    except ImportError:
        # distutils not available (thanks distros), use a concise poor
        # man's version parser.
        class Version(tuple):
            def __new__(cls, version_str):
                def parseint(s):
                        return int(s)
                    except ValueError:
                        return 0
                return tuple.__new__(cls, [parseint(s) for s in version_str.split('.')])

    import pathlib
    import tempfile
    import time
    cache = pathlib.Path(tempfile.gettempdir()) / 'googler-latest-version'
    latest_version_str = None
    # Try to load latest version string from cached location, if it
    # exists and is fresh enough.
        if cache.is_file() and time.time() - cache.stat().st_mtime < 86400:
            latest_version_str = cache.read_text().strip()
    except OSError:
    if not latest_version_str:
            latest_version_str = get_latest_ref().lstrip('v')
        except Exception:
    if not latest_version_str:
    # Try to fetch latest version string from GitHub.
        current_version = Version(_VERSION_)
        latest_version = Version(latest_version_str)
    except ValueError:
    if latest_version > current_version:
        print('\x1b[33;1mThe latest release of googler is v%s, please upgrade.\x1b[0m'
              % latest_version_str,

# Miscellaneous functions

def python_version():
    return '%d.%d.%d' % sys.version_info[:3]

def https_proxy_from_environment():
    return os.getenv('https_proxy')

def parse_proxy_spec(proxyspec):
    if '://' in proxyspec:
        pos = proxyspec.find('://')
        scheme = proxyspec[:pos]
        proxyspec = proxyspec[pos+3:]
        if scheme.lower() != 'http':
            # Only support HTTP proxies.
            # In particular, we don't support HTTPS proxies since we
            # only speak plain HTTP to the proxy server, so don't give
            # users a false sense of security.
            raise NotImplementedError('Unsupported proxy scheme %s.' % scheme)

    if '@' in proxyspec:
        pos = proxyspec.find('@')
        user_passwd = urllib.parse.unquote(proxyspec[:pos])
        # Remove trailing '/' if any
        host_port = proxyspec[pos+1:].rstrip('/')
        user_passwd = None
        host_port = proxyspec.rstrip('/')

    if ':' not in host_port:
        # Use port 1080 as default, following curl.
        host_port += ':1080'

    return user_passwd, host_port

def set_win_console_mode():
    # VT100 control sequences are supported on Windows 10 Anniversary Update and later.
    # https://docs.microsoft.com/en-us/windows/console/console-virtual-terminal-sequences
    # https://docs.microsoft.com/en-us/windows/console/setconsolemode
    if platform.release() == '10':
        STD_OUTPUT_HANDLE = -11
        STD_ERROR_HANDLE = -12
            from ctypes import windll, wintypes, byref
            kernel32 = windll.kernel32
            for nhandle in (STD_OUTPUT_HANDLE, STD_ERROR_HANDLE):
                handle = kernel32.GetStdHandle(nhandle)
                old_mode = wintypes.DWORD()
                if not kernel32.GetConsoleMode(handle, byref(old_mode)):
                    raise RuntimeError('GetConsoleMode failed')
                new_mode = old_mode.value | ENABLE_VIRTUAL_TERMINAL_PROCESSING
                if not kernel32.SetConsoleMode(handle, new_mode):
                    raise RuntimeError('SetConsoleMode failed')
            # Note: No need to restore at exit. SetConsoleMode seems to
            # be limited to the calling process.
        except Exception:

# Query autocompleter

# This function is largely experimental and could raise any exception;
# you should be prepared to catch anything. When it works though, it
# returns a list of strings the prefix could autocomplete to (however,
# it is not guaranteed that they start with the specified prefix; for
# instance, they won't if the specified prefix ends in a punctuation
# mark.)
def completer_fetch_completions(prefix):
    import html
    import json
    import re
    import urllib.request

    # One can pass the 'hl' query param to specify the language. We
    # ignore that for now.
    api_url = ('https://www.google.com/complete/search?client=psy-ab&q=%s' %
               urllib.parse.quote(prefix, safe=''))
    # A timeout of 3 seconds seems to be overly generous already.
    resp = urllib.request.urlopen(api_url, timeout=3)
    charset = resp.headers.get_content_charset()
    logger.debug('Completions charset: %s', charset)
    respobj = json.loads(resp.read().decode(charset))

    # The response object, once parsed as JSON, should look like
    # ['git',
    #  [['git<b>hub</b>', 0],
    #   ['git', 0],
    #   ['git<b>lab</b>', 0],
    #   ['git<b> stash</b>', 0]],
    #  {'q': 'oooAhRzoChqNmMbNaaDKXk1YY4k', 't': {'bpc': False, 'tlw': False}}]
    # Note the each result entry need not have two members; e.g., for
    # 'gi', there is an entry ['gi<b>f</b>', 0, [131]].
    HTML_TAG = re.compile(r'<[^>]+>')
    return [html.unescape(HTML_TAG.sub('', entry[0])) for entry in respobj[1]]

def completer_run(prefix):
    if prefix:
        completions = completer_fetch_completions(prefix)
        if completions:

def parse_args(args=None, namespace=None):
    """Parse googler arguments/options.

    args : list, optional
        Arguments to parse. Default is ``sys.argv``.
    namespace : argparse.Namespace
        Namespace to write to. Default is a new namespace.

        Namespace with parsed arguments / options.


    colorstr_env = os.getenv('GOOGLER_COLORS')

    argparser = GooglerArgumentParser(description='Google from the command-line.')
    addarg = argparser.add_argument
    addarg('-s', '--start', type=argparser.nonnegative_int, default=0,
           metavar='N', help='start at the Nth result')
    addarg('-n', '--count', dest='num', type=argparser.positive_int,
           default=10, metavar='N', help='show N results (default 10)')
    addarg('-N', '--news', action='store_true',
           help='show results from news section')
    addarg('-V', '--videos', action='store_true',
           help='show results from videos section')
    addarg('-c', '--tld', metavar='TLD',
           help="""country-specific search with top-level domain .TLD, e.g., 'in'
           for India""")
    addarg('-l', '--lang', metavar='LANG', help='display in language LANG')
    addarg('-g', '--geoloc', metavar='CC',
           help="""country-specific geolocation search with country code CC, e.g.
           'in' for India. Country codes are the same as top-level domains""")
    addarg('-x', '--exact', action='store_true',
           help='disable automatic spelling correction')
    addarg('--colorize', nargs='?', choices=['auto', 'always', 'never'],
           const='always', default='auto',
           help="""whether to colorize output; defaults to 'auto', which enables
           color when stdout is a tty device; using --colorize without an argument
           is equivalent to --colorize=always""")
    addarg('-C', '--nocolor', action='store_true',
           help='equivalent to --colorize=never')
    addarg('--colors', dest='colorstr', type=argparser.is_colorstr,
           default=colorstr_env if colorstr_env else 'GKlgxy', metavar='COLORS',
           help='set output colors (see man page for details)')
    addarg('-j', '--first', '--lucky', dest='lucky', action='store_true',
           help='open the first result in web browser and exit')
    addarg('-t', '--time', dest='duration', type=argparser.is_duration,
           metavar='dN', help='time limit search '
           '[h5 (5 hrs), d5 (5 days), w5 (5 weeks), m5 (5 months), y5 (5 years)]')
    addarg('--from', type=argparser.is_date,
           help="""starting date/month/year of date range; must use American date
           format with slashes, e.g., 2/24/2020, 2/2020, 2020; can be used in
           conjunction with --to, and overrides -t, --time""")
    addarg('--to', type=argparser.is_date,
           help='ending date/month/year of date range; see --from')
    addarg('-w', '--site', dest='sites', action='append', metavar='SITE',
           help='search a site using Google')
    addarg('-e', '--exclude', dest='exclude', action='append', metavar='SITE',
           help='exclude site from results')
    addarg('--unfilter', action='store_true', help='do not omit similar results')
    addarg('-p', '--proxy', default=https_proxy_from_environment(),
           help="""tunnel traffic through an HTTP proxy;
           PROXY is of the form [http://][user:password@]proxyhost[:port]""")
    addarg('--noua', action='store_true', help=argparse.SUPPRESS)
    addarg('--notweak', action='store_true',
           help='disable TCP optimizations and forced TLS 1.2')
    addarg('--json', action='store_true',
           help='output in JSON format; implies --noprompt')
    addarg('--url-handler', metavar='UTIL',
           help='custom script or cli utility to open results')
    addarg('--show-browser-logs', action='store_true',
           help='do not suppress browser output (stdout and stderr)')
    addarg('--np', '--noprompt', dest='noninteractive', action='store_true',
           help='search and exit, do not prompt')
    addarg('-4', '--ipv4', action='store_const', dest='address_family',
           const=socket.AF_INET, default=0,
           help="""only connect over IPv4
           (by default, IPv4 is preferred but IPv6 is used as a fallback)""")
    addarg('-6', '--ipv6', action='store_const', dest='address_family',
           const=socket.AF_INET6, default=0,
           help='only connect over IPv6')
    addarg('keywords', nargs='*', metavar='KEYWORD', help='search keywords')
    if ENABLE_SELF_UPGRADE_MECHANISM and not system_is_windows():
        addarg('-u', '--upgrade', action='store_true',
               help='perform in-place self-upgrade')
        addarg('--include-git', action='store_true',
               help='when used with --upgrade, get latest git master')
    addarg('-v', '--version', action='version', version=_VERSION_)
    addarg('-d', '--debug', action='store_true', help='enable debugging')
    # Hidden option for interacting with DOM in an IPython/pdb shell
    addarg('-D', '--debugger', action='store_true', help=argparse.SUPPRESS)
    # Hidden option for parsing dumped HTML
    addarg('--parse', dest='html_file', help=argparse.SUPPRESS)
    addarg('--complete', help=argparse.SUPPRESS)

    parsed = argparser.parse_args(args, namespace)
    if parsed.nocolor:
        parsed.colorize = 'never'

    return parsed

def main():
        opts = parse_args()

        # Set logging level
        if opts.debug:
            logger.debug('googler version %s', _VERSION_)
            logger.debug('Python version %s', python_version())
            logger.debug('Platform: %s', platform.platform())

        if opts.debugger:
            global debugger
            debugger = True

        # Handle query completer
        if opts.complete is not None:

        # Handle self-upgrade
        if hasattr(opts, 'upgrade') and opts.upgrade:


        if opts.keywords:
                # Add cmdline args to readline history
                readline.add_history(' '.join(opts.keywords))
            except Exception:

        # Set colors
        if opts.colorize == 'always':
            colorize = True
        elif opts.colorize == 'auto':
            colorize = sys.stdout.isatty()
        else:  # opts.colorize == 'never'
            colorize = False

        if colorize:
            colors = Colors(*[COLORMAP[c] for c in opts.colorstr], reset=COLORMAP['x'])
            colors = None
        Result.colors = colors
        Result.urlexpand = True if os.getenv('DISABLE_URL_EXPANSION') is None else False
        GooglerCmd.colors = colors

        # Try to enable ANSI color support in cmd or PowerShell on Windows 10
        if sys.platform == 'win32' and sys.stdout.isatty() and colorize:

        if opts.url_handler is not None:
            open_url.url_handler = opts.url_handler
            # Set text browser override to False
            open_url.override_text_browser = False

            # Handle browser output suppression
            if opts.show_browser_logs or (os.getenv('BROWSER') in text_browsers):
                open_url.suppress_browser_output = False
                open_url.suppress_browser_output = True

        if opts.noua:
            logger.warning('--noua option has been deprecated and has no effect (see #284)')

        repl = GooglerCmd(opts)

        # Non-interactive mode
        if opts.json or opts.lucky or opts.noninteractive or opts.html_file:
            if opts.lucky:
                if repl.results:
                    print('No results.', file=sys.stderr)

        # Interactive mode
    except Exception as e:
        # With debugging on, let the exception through for a traceback;
        # otherwise, only print the exception error message.
        if logger.isEnabledFor(logging.DEBUG):

if __name__ == '__main__':

Log in or click on link to see number of positives.

In cases where actual malware is found, the packages are subject to removal. Software sometimes has false positives. Moderators do not necessarily validate the safety of the underlying software, only that a package retrieves software from the official distribution point and/or validate embedded software against official distribution point (where distribution rights allow redistribution).

Chocolatey Pro provides runtime protection from possible malware.

Add to Builder Version Downloads Last Updated Status
googler 4.3.1 310 Sunday, October 11, 2020 Approved
googler 4.2 238 Tuesday, July 28, 2020 Approved
googler 4.1 222 Wednesday, May 27, 2020 Approved
googler 3.9 634 Thursday, May 30, 2019 Approved
googler 3.8 326 Wednesday, March 27, 2019 Approved
googler 3.7.1 363 Tuesday, November 13, 2018 Approved

Discussion for the googler Package

Ground Rules:

  • This discussion is only about googler and the googler package. If you have feedback for Chocolatey, please contact the Google Group.
  • This discussion will carry over multiple versions. If you have a comment about a particular version, please note that in your comments.
  • The maintainers of this Chocolatey Package will be notified about new comments that are posted to this Disqus thread, however, it is NOT a guarantee that you will get a response. If you do not hear back from the maintainers after posting a message below, please follow up by using the link on the left side of this page or follow this link to contact maintainers. If you still hear nothing back, please follow the package triage process.
  • Tell us what you love about the package or googler, or tell us what needs improvement.
  • Share your experiences with the package, or extra configuration or gotchas that you've found.
  • If you use a url, the comment will be flagged for moderation until you've been whitelisted. Disqus moderated comments are approved on a weekly schedule if not sooner. It could take between 1-5 days for your comment to show up.
comments powered by Disqus