Browse Source

initial commit, separated from transient

Alberto G. Corona 6 years ago
  1. 24
  2. 674
  3. 679
  4. 2
  5. 4
  6. 5
  7. 46
  8. 115
  9. 56
  10. 158
  11. 141
  12. BIN
  13. 0
  14. 386
  15. 1268
  16. 156
  17. 125
  18. 5
  19. 65
  20. 72
  21. 54
  22. 87
  23. 13
  24. 47
  25. 11
  26. 55
  27. 127
  28. 74
  29. 67
  30. 13


@ -0,0 +1,24 @@
FROM agocorona/herokughcjs
RUN git clone transgit \
&& cd transgit \
&& git checkout ghcjs \
&& cabal install
RUN cd transgit && cabal install --ghcjs
RUN git clone \
&& cd ghcjs-perch \
&& cabal install \
&& cabal install --ghcjs #
RUN git clone \
&& cd ghcjs-hplay \
&& cabal install --ghcjs \
&& cabal install
ADD . /transient/
CMD cd /transient && chmod 777 && ./


@ -0,0 +1,674 @@
Version 3, 29 June 2007
Copyright (C) 2007 Free Software Foundation, Inc. <>
Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
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.
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.
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 <>.
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


@ -0,0 +1,679 @@
New: GHCJS integration and improved map-reduce (and wormholes and teleporting of computations)
Browser nodes, running transient programs compiled with ghcjs are
integrated with server nodes using websockets communnications.
Just compile the program with ghcjs and point the browser to http://server:port.
The server nodes have a HTTP server that will send the compiled program to
the browser.
Browser nodes integrate Hplayground (package ghcjs-hplay) they can create
widgets and control the server nodes. A computation can move from browser
to server and back at runtime despite the different architecture.
Widgets with code running in browser and servers cancompose with other widgets.
A Browser node can access to many server nodes using `clustered` primitives.
map-reduce (Transient.DDS module) now has a true shuffle stage. Before the final reduction was in the node that initiated the computation. Now it is also distributed. Look for the concept of shuffle in the spark cloud platform
`teleport` is a new primitive that translates computations back and forth reusing an
already opened connection.
The connection is initiated by `wormhole` which opens a connection with another node anywhere in a computation.
Don't worry: as always, everything is composable. All the previous distributed primitives are rewritten in terms of these two new ones.
How to run the new example with ghcjs code:
You nee ghc and ghcjs installed.
clone and install perch:
> git clone
> cd ghcjs-perch
> cabal install --ghcjs -f ghcjs
clone and install this branch (ghcjs) of transient:
> git clone
> cd transient
> git checkout ghcjs
clone and install hplay:
> git clone
> cd ghcjs-hplay
> cabal install
> cabal install --ghcjs -f ghcjs
for fast development interactions, use the script
> buildrun examples/webapp.hs
This will compile examples/webapp.hs for ghcjs and run it interpreted with runghc
then point a browser to: http:localhost:2020
See this video to see this example running:
The test program run among other things, two copies of a widget that start, stop and display a counter that run in the server.
One of the biggest dreams of software engineering is unrestricted composability.
This may be put in these terms:
let `ap1` and `ap2` two applications with arbitrary complexity, with all effects including multiple threads, asynchronous IO, indeterminism, events and perhaps, distributed computing.
Then the combinations:
- ap1 <|> ap2 -- Alternative expression
- ap1 >>= \x -> ap2 -- monadic sequence
- ap1 <> ap2 -- monoidal expression
- (,) <$> ap1 <*> ap2 -- Applicative expression
are possible if the types match, and generate new applications that are composable as well.
Transient does exactly that.
The operators `<|>` and `<>` can be used for concurrency, the operator `<|>` can be used for parallelism and `>>=` for sequencing of threads and/or distributed processes. So even in the presence of these effects and others, everything is composable.
For this purpose transient is an extensible effects monad with all major effects and primitives for parallelism, events, asyncronous IO, early termination, non-determinism logging and distributed computing. Since it is possible to extend it with more effects without adding monad transformers, the composability is assured.
The [Wiki]( is more user oriented
The articles are more tecnical:
- [Philosophy, async, parallelism, thread control, events, Session state](
- [Backtracking and undoing IO transactions](
- [Non-deterministic list like processing, multithreading](
- [Distributed computing](
- [Publish-Subscribe variables](
- [Distributed streaming, map-reduce](
These articles contain executable examples (not now, since the site no longer support the execution of haskell snippets).
This is the text of the first article:
# Introduction #
I have a problem: How can I present the few applicative and monadic combinators that I just developed. I could present it as
- something for multithreaded event handling without inversion of control. Or
- something for parallelization of processes: [async]( without the wait
- for automatic thread control
- for alternative and applicative composition of parallel IO actions
- for indeterminism and asynchronicity effects
- for high level programming at the specification level
- for creating and composing applications by means of a single expression
- for overcoming futures and promises of Scala and JavaScript making them unnecessary.
Too much stuff for a single article. Maybe I should split this article and take time to write something more extensive and less dense. But I'm lazy and moreover they are only a few primitives, four or five, and no new operators. Breaking the article would hide the big picture. It would not display the beautiful unity of the common solution.
- If you are interested in how the idea came about read the next paragraph
- If you are interested in the internals, read the section "Enter the monad"
- If you are interested in how the monad controls multiple threads see "Implicit thread control"
- If you are interested in examples read from "Example" on
- If you are interested in Async, promises and futures, read "Beyond futures and promises"
- If you are interested in running the examples read "Composition of Programs"
- If you are interested in de-inverting the control of callbacks see "deinverting callbacks"
#The problem: parallelization, concurrency and inversion of control#
Suppose that I have a blocking computation that returns data when something happens. It may be also a long running computation that blocks the thread for a time:
``` haskell
receive: IO a
I can use it as such, but it blocks. I cannot use it in a context where other events are firing. I must create a thread for each blocking IO call. All of these threads probably modify a central state. Otherwise there will be no communication of data among threads. Alternatively someone may have created a kind of framework for this particular problem, where these blocking calls are managed. It may be a GUI toolkit, or a Web application framework, A browser environment or a library for the management of an ATM machine etc. In any case, what the programmer sees is a set of blocking synchronous calls and a set of callbacks or handlers that he has to program and configure in the framework.
Blocking IO creates the need to resort to manual thread management and concurrency. That means that the code is split into parallel and concurrent chunks which are hard to code and debug. In the second case I have non blocking IO, since the thread management is done implicitly by the framework, but, on the other side, I have to split the program logic into disconnected chunks. As a result, the program logic is very hard to grasp. This is known as the callback hell, a consequence of the inversion of control.
#The OOP non-solution#
The second scenario appears when threading is managed by a framework. Essentially it is the same case. In both cases we end up with disconnected chunks of code and a mutable state. The standard way to manage this messy central state has been to divide it into smaller states and encapsulate them with the methods that modify and serve the state. This is the Object Oriented Programming solution; The first OOP languages were created for managing events (SIMULA), and mouse events in an interactive GUI (Smalltalk).
*Object-oriented programming is an exceptionally bad idea which could only have originated in California. - Dijkstra*
Object Oriented Programming naturally fits with this inversion of control, that pervades IT problems. Whenever there are more than one asynchronous input, there is multitasking or inversion of control with state. The solution is a state machine. What OOP does is to split this state machine into smaller state machines called objects, that interact among them. But that implies the need to "deconstruct" the specifications.
##Deconstruct the specification recipe considered harmful##
Usually the specification of something is naturally expressed as if this "something" is a process, in the third person active perspective. People describe things that are the focus of their analysis as active, and the environment is described as passive even if there are active parts is the environment. In a recipe they say: you "fry the eggs" not "there is fire and there is eggs, you start the fire and the eggs are fried by the fire". You see that the passive perspective implies to give protagonism to low level elements that you are not interested in when writing an specification. Creating an OOP solution implies the deconstruction of the specification recipe into multiple third person passive perspectives, one for each class or object.
In any case, the programs are made of disconnected pieces, the software does not follow the natural flow as it would naturally extracted from the specifications and has many low level details, so the resulting code is low level, it is hard to maintain and the final service created is not composable. There is no way to insert your service within something bigger with a single invocation. This derives in the scarce reusability of the software, and the need of profuse documentation. It is inelegant, buggy, hard to maintain, and permits an huge number of arbitrary alternatives in the design space that aggravates the mentioned problems. It may be though that this is good for the IT business, but not in the medium-long term. And moreover, there are better ways to do it.
Functional languages help little on that if people limit themselves to porting already existing solutions from OOP languages.
#What we need#
*Simplicity is prerequisite for reliability. - Dijkstra*
The application must be programmed following the natural flow defined in the specification. The code must not split the specifications into parallel running tasks, neither invert the control and deconstruct the specification into objects. The design space must be limited so everyone should program the same specification in the same way. So other's code can be grasped immediately without the aid of external documentation. The application must transport user-defined state, that can be inspected and updated, added and deleted, but this state is instrumental it is not the center, because the center is the process described in the specification.
**We need an EDSL for hardworking IT programmers, that use Java, JavaScript, Scala, C#, PHP, Ruby or Python and don't know Haskell. Not a monad stack but a simple Monad that may liberate them from the Oppressive Object Paradigm, or OOP, without forcing them to sacrifice to the gods of Category Theory. With applicative and alternative combinators and a few primitives for implicit parallelization and thread control and for de-inversion of callbacks in the IO monad. Plus user-defined state management and early termination**
*“Simplicity is a great virtue but it requires hard work to achieve it and education to appreciate it. And to make matters worse: complexity sells. - Dijkstra*
#Enter the monad#
A monad with the asynchronicity effect can rescue the IT industry from the inversion of control trap for which OOP was originally designed while allowing implicit parallelization and thread control. A entire application can be coded in a single monadic expression with little or no plumbing. That allows the creation of composable applications and services of the [A -> A -> A]( kind.
In [A monad for reactive programming]( I defined a monad that de-inverts the control when there are different events. The `Transient` monad can listen for events at different points in the monadic expression. It is no longer necessary to have a single listen point like in the case of the event loop or in classical reactive frameworks. Moreover, these event listeners do not block, so every event watchig point is active in the monad at the same time.
The events in the above article are injected by a simulated event loop in the state monad. This time I will show how to listen for IO computations without the help of a framework that brings events. These events may be hardware buttons, device driver inputs, requests from users, responses from databases or requests from other systems in the cloud, etc.
What we intend here is to formulate a general solution that permits coding close to the user requirement document and presenting the application as a single process even if involves multiple inputs and parallel executions in a single monadic expression. This expression will automatically spawn, communicate and kill tasks whenever necessary. We will see that we can improve the readability and reduce the complexity, so we can increase the maintainability and make entire services or applications composable.
Since I have to deal with dirty things like blocking, threads and IO, don't expect what follows to be a walk in the Platonic realm. I start with a monad like the `Transient` monad, that can be stopped with `empty` and continued with `runCont cont` where `cont` is the continuation context, set with `getCont`. (see explanation below)
``` haskell
data Transient m x= Transient {runTrans :: m (Maybe x)}
data EventF = forall a b . EventF
{xcomp :: (TransientIO a)
,fcomp :: [a -> TransientIO b]
, ... other ....}
type StateIO= StateT EventF IO
type TransientIO= Transient StateIO
instance Monad TransientIO where
return x = Transient $ return $ Just x
x >>= f = Transient $ do
cont <- setEventCont x f
mk <- runTrans x
resetEventCont cont
case mk of
Just k -> runTrans $ f k
Nothing -> return Nothing
instance Applicative TransientIO where
pure a = Transient . return $ Just a
Transient f <*> Transient g= Transient $ do
k <- f
x <- g
return $ k <*> x
instance Alternative TransientIO where
empty= Transient $ return Nothing
Transient f <|> Transient g= Transient $ do
k <- f
x <- g
return $ k <|> x
getCont ::(MonadState EventF m) => m EventF
getCont = get
runCont :: EventF -> StateIO ()
runCont (EventF x fs ...)= do runIt x (unsafeCoerce fs); return ()
runIt x fs= runTrans $ x >>= compose fs
compose []= const empty
compose (f: fs)= \x -> f x >>= compose fs
For a view of how this monad has evolved, look at the first article [A monad for reactive programming part 1]( where I present a simpler version of this monad that has some shortcomings. In the second part I solved these shortcomings. I think that this is the best way to understand it.
What this monad does is to store the closure `x` and the continuations `f` in the state. `getCont` captures the execution state at the point and `runCont` executes it.
As far as "continuation" is taken here, there may be more than one of them.
For example, in this expression:
x0 >>=((x >>= f1) >>= f2) >>= f3
for the closure generated at the execution point `x`, the continuations are
f1 >>= f2 >>= f3
and the closure is the result of the execution of `x0 >>= x`
What `setEventCont` and `resetEventCont` do is to compose the list of continuations (one for each nested expression) in a 'flattened' representation, as a list in `fcomp`. Since the list does not "know" that the continuations types match, I have to erase the types using `unsafeCoerce`.
# Parallelization #
With these three primitives `getCont`, `runCont` and `empty` I will define a `async` primitive that will run a blocking IO action in a new thread and will execute the continuation in that thread whenever in receives something:
``` haskell
buffer :: Dynamic
buffer= unsafePerformIO $ newEmptyMVar
async :: IO a -> TransientIO a
async receive = do
cont <- getCont
r <- liftIO $ tryTakeMVar buffer
case r of
Nothing ->do
liftIO . forkIO $ do
r <- receive
putMVar buffer $ toDync r
runCont cont
Just r -> return $ formDynamic r
Essentially, `async` gets the continuation, then inspects the buffer. If there is `Nothing` then spawn `receive` in a new thread. The current thread is finished (`empty`). When something arrives, it is put in the buffer, then `runCont` will continue at the beginning of `receive` in the new thread. It does so because `getCont` got the `Transient` continuation there. This time, there will be something in the buffer and will return it, so the procedure will continue after the event arrives, but in a new thread.
Note that `receive` only fills the buffer. When `runCont` executes the closure, it will inspect the buffer again. This time there will be something, the closure will succeed and the continuation will fire.
`getCont` and `runCont` are similar to [setjmp and longjmp]( in C. Moreover, the mechanism is not very different form the IO scheduler in GHC or in any operating system. But this time it runs at the application level rather than at the GHC level.
#Wait for events#
If we want to trigger the continuation repeatedly whenever something is received by `receive`, it is a matter of adding a loop to the `Nothing` branch. Then the continuation will be called for every received event.
Let's call this variant `waitEvents`:
``` haskell
waitEvents :: IO a -> TransientIO a
waitEvents receive = do
cont <- getCont
r <- tryTakeMVar buffer
case r of
Nothing ->do
liftIO . forkIO $ loop $ do
r <- receive
putMVar buffer r
runCont cont
Just r -> return r
loop x= x >> loop x
This program will say "hello" to every name entered.
``` haskell
runTransient :: TransientIO x -> IO (Maybe x, EventF)
runTransient t= runStateT (runTrans t) eventf0
main= do
runTransient $ do
name <- waitEvents getLine
liftIO $ putStrLn $ "hello "++ name
Note that there is no loop. `waitEvent` install `getLine` at the start of a process that execute the continuation, what is after `getLine`, for each entry. The loop is internal to `waitEvents`
Here `runTransient` executes a transient computation down to the IO monad.
`stay` is whatever that keeps the console application from exiting. That is because since the transient branch that waits for events is non-blocking, it would finish immediately. **After `async` or `waitEvents`, the current thread dies and the rest of the monadic computation runs in a different thread**
#Implicit thread control#
Since each event in any part of the monadic computation is active and triggers the continuation of the monad at that point, the monadic expression is multithreaded and non deterministic.
How to control the threads? It is natural to think that since `waitEvents` and `async` execute continuations within the monadic expression, then once something happens in a statement then their continuations must be invalidated.
That means that whenever `async` of `waitEvents` receive something, the threads that are running below must be killed. Then this statement, with the new buffered input will execute his closure and rebuild the continuation again.
This is the natural thread management that I implemented. I do not detail the modifications necessary for `waitEvents` to permit this behaviour. It is a matter of keeping in the state the list of spawned threads so that each `waitEvents` has the information about all the threads that are triggered after it. Additionally, this list contains also a buffer for each of these threads.
In this example:
``` haskell
main= do
runTransient $ do
waitEvents watchReset <|> return ()
name <- waitEvents getLine
liftIO $ putStrLn $ "hello "++ name
the `return()` composed with the alternative operator `<|>` would bypass immediately the wait for the reset event, but as soon as the reset is pressed, all the event handlers spawned after it will be killed. Immediately they will be spawned again.
This is a slightly different version:
main= do
runTransient $ do
r <- (waitEvents watchStop >> return True) <|> return False
if r then liftIO $ putStrln "STOP" else do
name <- waitEvents getLine
liftIO $ putStrLn $ "hello "++ name
In this case the program will be stopped and will not be re-spawned when `watchStop` is activated. Since now the branch of the monad executed is different. It prints the stop message and finalizes.
#Non blocking IO#
Let's create a nonblocking keyboard input thing called `option`. At the same time this is a good example of inter-thread communication within the `Transient` monad:
``` haskell
option :: (Typeable a, Show a, Read a, Eq a) =>
a -> [Char] -> TransientIO a
option ret message= do
liftIO $ putStrLn $ message++"("++show ret++")"
waitEvents "" getLine'
getLine'= do
atomically $ do
mr <- readTVar getLineRef
case mr of
Nothing -> retry
Just r ->
case readsPrec 0 r of
[] -> retry
(s,_):_ -> if ret== s
then do
writeTVar getLineRef Nothing
return ret
else retry
_ -> retry
getLineRef= unsafePerformIO $ newTVarIO Nothing
inputLoop :: IO ()
inputLoop= do
r<- getLine !> "started inputLoop"
if r=="end" return True else do
atomically . writeTVar getLineRef $ Just r
# Applicative and Alternative combinators#
`option` reads the standard input in nonblocking mode, so that many options can be combined using applicative or alternative operators. `option` shows a message and waits for `inputLoop` to enter an input line. If some `option` match, it returns the value. If it does not match, it fails with `empty`, but the loop in `waitEvents` re-executes `getLine'` again for this option. In this way, the options are continuously watching the input. Note that more than one option can be triggered simultaneously, in a different thread.
`inputLoop` is initialized by `async`. It waits for input, and exposes it to all the running `getLine'` processes (one per `option`) in a `TVar`. If the user presses "end" `inputLoop` returns and `async` kills all the watching threads below.
``` haskell
main= do
runTransient choose
choose :: TransientIO()
choose= do
r <- async inputLoop <|> return False
case r of
True -> return ()
False-> do
r <- option (1 :: Int) "red" <|> option 2 "green" <|> option 3 "blue"
liftIO $ print r
The above program will print repeatedly the chosen option. We see that `option` is composable using the alternative operator.
Now let's create another event generator, a number is sent every second, while two options are waiting for keyboard input:
``` haskell
data Option= Option String String | Number Int deriving Show
choose= do
r <- ( Option <$> ( option "1" "red" <*> option "2" "green"))
<|> ( Number <$> waitEvents waitnumber )
liftIO $ putStrLn $ "result=" ++ show r
waitnumber= do
threadDelay 1000000
return 42
Applicative and alternative combinators can be used fully. The Applicative waits for both events to be triggered to have data in their respective buffers. `waitnumber` produces an event each second.
Each Option runs a different `waitEvent` in a different thread, but the closure is the same applicative expression in the three threads. The three have a TVar waiting for new input. They fill their respective buffers when they validate. The thread that fills the last buffer empties them and executes the continuation. The other two fail, but stay ready for the next input, since `waitEvents` has a loop.
#Beyond futures and promises#
Scala Futures and the haskell library async manage placeholders that receive the result and place them where the result of the computation is needed.
Scala Futures also uses futures in nice chains of multi-threaded lists that can be transformed in the style of map-reduce.
In this sense they are similar to the javaScript promises, which chain code with `then`, but the latter does not perform multiple tasks like in the case of Scala futures.
For some needs Scala and JavaScript must use callbacks since the constraints of their frameworks do not allow enough flexibility. futures and promises force the programmer to enter in a different kind of computation model, different from the one of the native languages. In the case of Scala it is monoidal. In the case of Javascript it is a restricted form of bind operation.
This library puts the continuation code at the end of the receiving pipeline and parallelizes the execution, but the continuation is the plain monadic code that is after the receiving call in the monadic expression, so there is no restriction about what can be done.
`asyn` can be used for any process that we want to parallelize. Now we can do it better than the [async]( library. This program sums the words in google and haskell homepages **in parallel**. Using [Network.HTTP](
``` haskell
sum= do
(r,r') <- (,) <$> async (worker "")
<*> async (worker "")
liftIO $ putStrLn $ "result=" ++ show r + r'
getURL= simpleHTTP . getRequest
worker :: String -> IO Int
worker url=do
r <- getURL url
body <- getResponseBody r
return . length . words $ body
That is a complete working example. Note that unlike in the async library, there is no `wait` primitive. All the processing is done in the `worker` in his own thread. When both workers finish, they return the result to the monad and continue the thread.
We also can do parallel IO processing in the style of futures of the Scala language using the `Monoid` instance of `TransientIO`. But this time since we use continuations, futures are no longer necessary: the last download triggers the continuation.
``` haskell
instance Monoid a => Monoid (TransientIO a) where
mappend x y = mappend <$> x <*> y
mempty= return mempty
sum= do
rs <- foldl (<>) (return 0) $ map (async . worker)
[ "", ""]
liftIO $ putStrLn $ "result=" ++ show rs
Since the `worker`returns an `Int`, to sum the results we need a `Monoid` instance for `Int`
``` haskell
instance Monoid Int where
mappend= (+)
mempty= 0
Note that there is a full de-inversion of control, since the result returns to the monad. The Futures and promises of Scala (or in javaScript) can not return the execution flow to the calling procedure. Since the Transient monad continues in the thread that completed the worker, the processing can continue in the monad. That permits more complex and yet clearer computations. It is not reduced to list-like processing.
# A Web Server #
Here is a Web Server:
``` haskell
server= do
sock <- liftIO $ listenOn $ PortNumber 80
(h,_,_) <- spawn $ accept sock
liftIO $ do
hPutStr h msg
hFlush h
hClose h
msg = "HTTP/1.0 200 OK\r\nContent-Length: 5\r\n\r\nPong!\r\n"
In the current code, the primitives `async`, `waitEvents` and `spawn` are defined in terms of `parallel`, which is a generalization of `async` and `waitEvents` explained above:
``` haskell
data Loop= Once | Loop | Multithread
waitEvents :: IO b -> TransientIO b
waitEvents= parallel Loop
async :: IO b -> TransientIO b
async = parallel Once
spawn= parallel Multithread
parallel :: Loop -> IO b -> TransientIO b
When `parallel` is called with `Multithread`, it spawns the continuation in a thread for each event received immediately without waiting for the termination of the previous one. `waitEvent` executes the continuation within the thread so the receive method is not called again until the previous event is processed.
# Composition of programs (Runnable example) #
We can compose any of these programs together since none of them blocks and the automatic thread control applies gracefully to all of the elements. This program combines the above programs and some others.
The combination in this case is using the alternative operator:
colors <|> app <|> sum1 <|> sum2 <|> server <|> menu
Since the fpcomplete environment uses ghci and it shares threads among snippets of code, I can run only one example in this article, and the composability of Transient is nice to show them all togeter.
To verify the multitasking press: `app` and then `colors`. `app` would start an iterative counter with an applicative expression, while `colors` will ask for an option among tree of them. Both will run in parallel until you press "main" which will stop both, since main is above in the monad.
Code running at the original article:
#Session data#
I added a type indexed map to the state so the user can store his own session data with these primitives:
``` haskell
setSData :: ( MonadState EventF m,Typeable a) => a -> m ()
getSData :: MonadState EventF m => Typeable a =>Transient m a
Session data can be used instead of a state monad transformer for each new kind of user data.
My purpose is to create a monad for general IT purposes, for profane programmers with no knowledge of monad transformers.
Since `empty` stops the computation but does not transport any error condition, session data can be used for this purpose:
``` haskell
data Status= Error String | NoError
fail :: String -> TransientIO a
fail msg= setSData (Error msg) >> empty
After the execution, I can inspect the status:
``` haskell
status <- getSData <|> NoError
The alternative expression is necessary since if `Status` has not been set, the computation would stop. `NoError` guarantees that it does not stop.
#de-inverting callbacks#
So far so good. But what happens when besides dealing with raw blocking IO there is a framework that deals with some particular events, so it initiates the threads himself and expects you just to set the callbacks?
Suppose that we have this event handling setter:
``` haskell
setHandlerForWhatever :: (a -> IO ()) -> IO ()
It is necessary a de-inversion call `whateverHappened` at some point of the computation (may be at the beginning) so that the callback continues the monadic execution:
``` haskell
r <- whatheverHappened
doSomethingWith r
To define the de-inverted call `whateverHappened` we use the same trick than in `async`, but this time there is no `forkIO` neither thread control, since the framework does it for you:
``` haskell
whateverHappened= do
cont <- getCont
mEvData <- Just <*> getSData <|> return Nothing
case mEvData of
Nothing -> setHandlerForWhatever $\dat -> do
runStateT ( setSData dat >> runTansient cont) cont
Just dat -> return dat
Whether the framework is single threaded or multi-threaded is not important, we give it the event handlers that it needs by means of continuations.
To have something more general, I defined:
``` haskell
type EventSetter eventdata response= (eventdata -> IO response) -> IO ()
type ToReturn response= IO response
:: Typeable eventdata
=> EventSetter eventdata response
-> ToReturn response
-> TransientIO eventdata
The second parameter is the value returned by the callback. So if you have a callback called `OnResponse`
``` haskell
data Control= SendMore | MoMore
OnResponse :: (Response -> IO Control) -> IO()
I can display all data received while controlling the reception this way:
``` haskell
rcontrol <- newMVar Control
resp <- react $ OnResponse (const $ readMVar rcontrol)
display resp
r <- (option "more" "more" >> return SendMore) <|> (option "stop" "stop" >> return NoMore)
putMVar rcontrol r
Since `react` set as callback all the rest of the computation and since the ToReturn expression is evaluated the latest, the continuation is executed and sets `rcontrol` before the `ToReturn` expression is evaluated.
Note that you can reassign the callback at any moment since react would set whatever continuation that is after it.
#Conclusions and future work#
The code is at
My aim is to create a family of combinators for programming in industry. As I said before, that implies no monad transformers, the simplest monad that could produce the simplest error messages.
The haskell applicative, alternative, monoidal and monadic combinators when applied to a monad that manage asynchronous IO permits multi-threaded programming with little plumbing that is close to the specification level with great composability. No inversion of control means no need to deconstruct the specifications and no state machines.
This, together with the uniform and composable thread management, narrows the design space and makes the application more understandable from the requirements, and thus the technical documentation and maintenance costs are reduced to a minimum.
Note that the bulk of the programming is done in the IO monad. That is on purpose. The idea is a simple IT EDSL with the rigth effects that permit rapid and intuitive development. Additional monads can be used by running them within the IO procedures defined by the programmer if they wish. I will add some additional effects like backtracking to undo transactions and to produce execution traces. That would be the base of a new version of MFlow, my server-side framework and integration platform. The ability to perform rollbacks and respond to asynchronous events at the same time is important for cloud applications. Reader- and Writer-effects for any programmer need are almost trivial to implement using `getSData` and `setSData`.
Resource allocation and deallocation for file handlers etc can be done using the same strategy used for thread control, but it is more orthogonal to delegate it to the IO threads themselves. The programmer can use exceptions or monads that guarantee proper release of resources before the thread is killed.
In `Multithread` mode the single entry buffer can be overrun. It is necessary to handle this case or, else, assume that the `receive` procedure has his own buffer and his own event contention mechanism. That is the most orthogonal option.
This is huge. I plan to create interfaces for some GUI toolkit. The GUI objects will be fully composable for the first time.
Spawning threads in other machines is the next big step. [MFlow]( and [hplayground]( will converge with this platform. If you want to collaborate, don't hesitate to send me a message!
With the `react` primitive it is possible to de-invert any framework, including the callbacks of a GUI toolkit, so the widgets can be managed with Applicative and monadic combinators. hplayground does that for the JavaScript callbacks and HTML forms. Since hplayground, that runs in the client and MFlow that runs on the server side share the same widget EDSL, that can be ported to a GUI, an application can run in any environment, including console applications.


@ -0,0 +1,2 @@
import Distribution.Simple
main = defaultMain


@ -0,0 +1,4 @@
ghcjs -isrc -i../ghcjs-hplay/src %1 -o static/out
if %errorlevel% neq 0 exit
runghc -isrc -i../ghcjs-hplay/src %1


@ -0,0 +1,5 @@
set -e
ghcjs -isrc -i../ghcjs-hplay/src $1 -o static/out
runghc -isrc -i../ghcjs-hplay/src $1


File diff suppressed because one or more lines are too long


@ -0,0 +1,115 @@
-- Distributed streaming using Transient
-- See the article:
{-# LANGUAGE ScopedTypeVariables, DeriveDataTypeable, MonadComprehensions #-}
module MainCountinuous where
import Transient.Base
import Transient.Move
import Transient.Indeterminism
import Transient.Logged
import Transient.Stream.Resource
import Control.Applicative
import Data.Char
import Control.Monad
import Control.Monad.IO.Class
import System.Random
import Data.IORef
import System.IO
import GHC.Conc
import System.Environment
-- continuos streaming version
-- Perform the same calculation but it does not stop, and the results are accumulated in in a mutable reference within the calling node,
-- so the precision in the value of pi is printed with more and more precision. every 1000 calculations.
-- Here instead of `collect` that finish the calculation when the number of samples has been reached, i use `group` which simply
-- group the number of results in a list and then sum of the list is returned to the calling node.
-- Since `group` do not finish the calculation, new sums are streamed from the nodes again and again.
main= do
let numNodes= 5
numCalcsNode= 100
ports= [2000.. 2000+ numNodes -1]
createLocalNode p= createNode "localhost" p
nodes= map createLocalNode ports
rresults <- newIORef (0,0)
keep $ freeThreads $ threads 10 $ runCloud $ do
--setBufSize 1024
local $ addNodes nodes
foldl (<|>) empty (map listen nodes) <|> return()
r <- clustered $ do
--Connection (Just (_,h,_,_)) _ <- getSData <|> error "no connection"
--liftIO $ hSetBuffering h $ BlockBuffering Nothing
r <- local $ group numCalcsNode $ do
n <- liftIO getNumCapabilities
threads n .
spawn $ do
x <- randomIO :: IO Double
y <- randomIO
return $ if x * x + y * y < 1 then 1 else (0 :: Int)
return $ sum r
(n,c) <- local $ liftIO $ atomicModifyIORef' rresults $ \(num, count) ->
let num' = num + r
count'= count + numCalcsNode
in ((num', count'),(num',count'))
when ( c `rem` 1000 ==0) $ local $ liftIO $ do
th <- myThreadId
putStrLn $ "Samples: "++ show c ++ " -> " ++
show( 4.0 * fromIntegral n / fromIntegral c)++ "\t" ++ show th
-- really distributed version
-- generate an executable with this main and invoke it as such:
-- program myport remotehost remoteport
-- where remotehost remoteport are from a previously initialized node
-- The first node initialize it with:
-- program myport localhost myport
mainDistributed= do
args <- getArgs
let localPort = read (args !! 0)
seedHost = read (args !! 1)
seedPort = read (args !! 2)
mynode = createNode "localhost" localPort
seedNode = createNode seedHost seedPort
numCalcsNode= 100
rresults <- liftIO $ newIORef (0,0)
runCloud' $ do
connect mynode seedNode
local $ option "start" "start the calculation once all the nodes have been started" :: Cloud String
r <- clustered $ do
--Connection (Just (_,h,_,_)) _ <- getSData <|> error "no connection"
--liftIO $ hSetBuffering h $ BlockBuffering Nothing
r <- local $ group numCalcsNode $ do
n <- liftIO getNumCapabilities
threads n .
spawn $ do
x <- randomIO :: IO Double
y <- randomIO
return $ if x * x + y * y < 1 then 1 else (0 :: Int)
return $ sum r
(n,c) <- local $ liftIO $ atomicModifyIORef' rresults $ \(num, count) ->
let num' = num + r
count'= count + numCalcsNode
in ((num', count'),(num',count'))
when ( c `rem` 1000 ==0) $ local $ liftIO $ do
th <- myThreadId
putStrLn $ "Samples: "++ show c ++ " -> " ++
show( 4.0 * fromIntegral n / fromIntegral c)++ "\t" ++ show th


@ -0,0 +1,56 @@
-- Distributed streaming using Transient
-- See the article:
{-# LANGUAGE ScopedTypeVariables, DeriveDataTypeable, MonadComprehensions #-}
module MainOnce where
import Transient.Base
import Transient.Move
import Transient.Indeterminism
import Transient.Logged
import Transient.Stream.Resource
import Control.Applicative
import Data.Char
import Control.Monad
import Control.Monad.IO.Class
import System.Random
import Data.IORef
import System.IO
import GHC.Conc
import System.Environment
-- distributed calculation of PI
-- This example program is the closest one to the defined in the spark examples:
-- But while the spark example does not contain the setup of the cluster and the confuguration/initalization
-- this examples includes everything
-- The nodes are simulated within the local process, but they communicate trough sockets and serialize data
-- just like real nodes. Each node spawn threads and return the result to the calling node.
-- when the number of result are reached `colect` kill the threads, the sockets are closed and the stream is stopped
-- for more details look at the article:
main= do
let numNodes= 5
numSamples= 1000
ports= [2000.. 2000 + numNodes -1]
createLocalNode p= createNode "localhost" p
nodes= map createLocalNode ports
keep $ do
addNodes nodes
-- option "start" "start"
xs <- collect numSamples $ runCloud $ do
foldl (<|>) empty (map listen nodes) <|> return()
local $ threads 2 $ runCloud $
clustered[if x * x + y * y < (1 :: Double) then 1 else (0 :: Int)| x <- random, y <-random]
liftIO $ print (4.0 * (fromIntegral $ sum xs) / (fromIntegral numSamples) :: Double)
random= local $ waitEvents' randomIO :: Cloud Double


@ -0,0 +1,158 @@
{-# LANGUAGE ScopedTypeVariables, DeriveDataTypeable, MonadComprehensions #-}
module Main where
import Transient.Move
import Transient.Logged
import Transient.Base
import Transient.Indeterminism
import Transient.EVars
import Network
import Control.Applicative
import Control.Monad.IO.Class
import System.Environment
import System.IO.Unsafe
import Data.Monoid
import System.IO
import Control.Monad
import Data.Maybe
import Control.Exception
import Control.Concurrent (threadDelay)
import Data.Typeable
import Data.IORef
import Data.List((\\))
-- to be executed with two or more nodes
main = do
args <- getArgs
if length args < 2
then do
putStrLn "The program need at least two parameters: localHost localPort remoteHost RemotePort"
putStrLn "Start one node alone. The rest, connected to this node."
return ()
else keep $ do
let localHost= head args
localPort= read $ args !! 1
(remoteHost,remotePort) =
if length args >=4
then(args !! 2, read $ args !! 3)
else (localHost,localPort)
let localNode= createNode localHost localPort
remoteNode= createNode remoteHost remotePort
connect localNode remoteNode
examples = do
logged $ option "main" "to see the menu" <|> return ""
r <- logged $ option "move" "move to another node"
<|> option "call" "call a function in another node"
<|> option "chat" "chat"
<|> option "netev" "events propagating trough the network"
case r of
"call" -> callExample
"move" -> moveExample
"chat" -> chat
"netev" -> networkEvents
data Environ= Environ (IORef String) deriving Typeable
callExample = do
node <- logged $ do
nodes <- getNodes
myNode <- getMyNode
return . head $ nodes \\ [myNode]
logged $ putStrLnhp node "asking for the remote data"
s <- callTo node $ do
putStrLnhp node "remote callTo request"
liftIO $ readIORef environ
liftIO $ putStrLn $ "resp=" ++ show s
{-# NOINLINE environ #-}
environ= unsafePerformIO $ newIORef "Not Changed"
moveExample = do
node <- logged $ do
nodes <- getNodes
myNode <- getMyNode
return . head $ nodes \\ [myNode]
putStrLnhp node "enter a string. It will be inserted in the other node by a migrating program"
name <- logged $ input (const True)
beamTo node
putStrLnhp node "moved!"
putStrLnhp node $ "inserting "++ name ++" as new data in this node"
liftIO $ writeIORef environ name
chat :: TransIO ()
chat = do
name <- logged $ do liftIO $ putStrLn "Name?" ; input (const True)
text <- logged $ waitEvents $ putStr ">" >> hFlush stdout >> getLine' (const True)
let line= name ++": "++ text
clustered $ liftIO $ putStrLn line
networkEvents = do
node <- logged $ do
nodes <- getNodes
myNode <- getMyNode
return . head $ nodes \\ [myNode]
logged $ putStrLnhp node "<- write \"fire\" in this other node"
r <- callTo node $ do
option "fire" "fire event"
return "event fired"
putStrLnhp node $ r ++ " in remote node"
putStrLnhp p msg= liftIO $ putStr (show p) >> putStr " ->" >> putStrLn msg
--call host port proc params= do
-- port <- getPort
-- listen port <|> return
-- parms <- logged $ return params
-- callTo host port proc parms
-- close
--distribute proc= do
-- case dataFor proc
-- Nothing -> proc
-- Just dataproc -> do
-- (h,p) <- bestNode dataproc
-- callTo h p proc
--bestNode dataproc=
-- nodes <- getNodes
-- (h,p) <- bestMatch dataproc nodes <- user defined
--bestMatch (DataProc nodesAccesed cpuLoad resourcesNeeded) nodes= do
-- nodesAccesed: node, response
--bestMove= do
-- thisproc <- gets myProc
-- case dataFor thisproc
-- Nothing -> return ()
-- Just dataproc -> do
-- (h,p) <- bestNode dataproc
-- moveTo h p
--inNetwork= do
-- p <- getPort
-- listen p <|> return ()


@ -0,0 +1,141 @@
module Main where
import Prelude hiding (div,id,span)
import Transient.Base
#ifdef ghcjs_HOST_OS
hiding ( option,runCloud')
import GHCJS.HPlay.View
#ifdef ghcjs_HOST_OS
hiding (map)
hiding (map, option,runCloud')
import Transient.Move
import Transient.Indeterminism
import Control.Applicative
import Control.Monad
import Data.Typeable
import Data.IORef
import Control.Concurrent (threadDelay)
import Control.Monad.IO.Class
-- Show the composability of transient web aplications
-- with three examples composed together, each one is a widget that execute
-- code in the browser AND the server.
main = simpleWebApp 2020 $ demo <|> demo2 <|> counters
notebook= do
r <- local . render $ textArea <*** wbutton () "send"
atServer $ do
-- execute with a pipe
demo= do
name <- local . render $ do
rawHtml $ do
p "this snippet captures the essence of this demonstration"
p $ span "it's a blend of server and browser code in a "
>> (span $ b "composable") >> span " piece"
div ! id (fs "fibs") $ i "Fibonacci numbers should appear here"
local . render $ wlink () (p " stream fibonacci numbers")
-- stream fibonancci
r <- atServer $ do
let fibs= 0 : 1 : zipWith (+) fibs (tail fibs) :: [Int] --fibonacci numb. definition
r <- local . threads 1 . choose $ take 10 fibs
lliftIO $ print r
lliftIO $ threadDelay 1000000
return r
local . render . at (fs "#fibs") Append $ rawHtml $ (h2 r)
demo2= do
name <- local . render $ do
rawHtml $ do
p "In this example you enter your name and the server will salute you"