DADA CLI is a CLI tool designed for testing the operation and features of Entra ID (Azure Active Directory). It enables the verification of functionalities in app registrations and enterprise applications, particularly focusing on SAML and OAuth 2.0.

• OAuth 2.0, Open ID Connect

  • You can experience the Authorization Code Flow and Client Credentials Flow.
  • Display and decode the obtained access tokens and ID tokens, enabling you to inspect their contents.
  • Use of simple Graph API operations to experience Continuous Access Evaluation (CAE).
  • Load a certificate file and create a JWT assertion.

  By utilizing these features, you can easily verify the information and functionalities included in the token claims within Entra ID.”

• SAML

  • You can easily experience SAML Single Sign-On (SSO) in Entra ID.
  • Generates SAML requests and decodes and displays SAML responses.
  • The command options allow you to specify the SAML request signature, Authentication Context, and Name ID Format.

  It allows you to easily test how Entra ID behaves when each of these settings is implemented.”

1. Installation

   $ git clone https://github.com/iamkdada/Azure-AD-OAuth-SAML-Python-Demo-CLI-APP.git
   $ cd Azure-AD-OAuth-SAML-Python-Demo-CLI-APP
   $ python3 -m venv venv
   $ pip3 install -r requirements.txt
   $ export PATH="$PATH:$PWD/src"

1. Download this project
2. Extract the downloaded project.
3. Create a virtual environment at this project dir.

   > python -m venv venv

4. Download the required libraries.

   > pip install -r requirements.txt

5. Set up the environment variables.

   PowerShell

   > $Env:PATH += ";$PWD\src"

   Command Prompt

   set PATH=%PATH%;%CD%\src

Plan to make improvements for easier installation.

1. Browse to [Azure Portal]>[Microsoft Entra ID]>[App Registrations] and select New registration.
2. Enter a Name for your application, for example dada-cli-oidc. Users of your app might see this name, and you can change it later.
3. Select bellow
   • Account Type : “Accounts in this organizational directory only”
   • Platform”Public client/native (mobile & desktop)”
   • Redirect uri : http://localhost
4. Select Register to create the application.

1. Setting Tenant ID & Client ID

   dada configure --tenant-id "<Your Tenant ID>" --client-id "Registered Application ID"

2. Let’s token request

   dada auth-code token-request

1. Browse to [Azure Portal]>[Microsoft Entra ID]>[Enterprise Application] and select New application.
2. Select Create your own application
3. Enter a Name for your application, for example dada-cli-saml. Users of your app might see this name, and you can change it later.
4. Select “Integrate any other application you don’t find in the gallery (Non-gallery)” and select Create
5. Browse to [Single sign-on]>[SAML] and select Edit.
6. Add identifier, for example dada.
7. Add reply URL as http://localhost

1. Setting Tenant ID & Client ID

   dada configure --tenant-id "<Your Tenant ID>" --entity-id "Registered Application Entity ID"

2. Let’s saml request

   dada saml saml-request

• auth code token request

  $ dada auth-code token-request
  "eyJ0eXAiOiJKV1QiLCJhbGciOiJSUzI1NiIsIng1dCI6Imk2bEdrM0ZaenhSY1ViMkMzbkV~~~~~~~~~
  "

• decode token

  $ dada auth-code show --token id --decode
  {
  "aud": "<GUID>",
  "exp": 1700021556,
  "iat": 1700017656,
  "iss": "https://login.microsoftonline.com/<tenant id>/v2.0",
  "name": "hoge hoge",
  "nbf": 1700017656,
  "oid": "<GUID>",
  "preferred_username": "hoge@*****.com",
  "rh": "0.AXwAji****************************",
  "sub": "XWFP_8f3rjEyjvlUzzTVB0v0W2I3DGxVn0*********",
  "tid": "GUID",
  "uti": "sBSSf-s2rkujr********",
  "ver": "2.0"
  }

• saml request

  $ dada saml saml-request --sign --force-authn
  "
  <decode saml resuponse>
  "

A simple dynamic charting project to fetch specific assets on Backend and render graphics with d3 on Frontend. Using Giraffe on Backend and Fable on Frontend. Full F# web application.

The project was develop to save the assets on server using Giraffe and provide the assets on the GET routes /silver and /gold. Before initialize the Giraffe server, a synchronous request is done to save locally SILVER.json and GOLD.json on /tmp folder. After that, two asynchronous task are initiated to download each file after 1 minute. So then the server giraffe is enable. This ensure that when /silver and /gold will be called at least one version of SILVER.json and GOLD.json exists on the server.

The frontend was designed to use Fable for transpiling F# |> JS and charting the graphics using Line Chart by binding the libs of d3.js and nvd3.js. Partially, some binds I got from an external repository: FableCharting

• dotnet SDK 1.0.4 and runtime
• node
• yarn
• Giraffe
• Fable
• Mono (for build: msbuild)

You need fetch the nodejs dependencies with yarn install on the root of the repository. So then build the Frontend and Backend, in that order. For building Frontend and Backend a shell script util is provided: build.sh.

The overall steps is:

yarn install
./build.sh

The build.sh will call dotnet restore && dotnet fable yarn-run build on the root of Frontend. This will generate a /public/bundle.js file as target. So then, will be called on src/Backend: dotnet clean; dotnet restore; msbuild; dotnet run. This should generate the binaries, copy the public folder as WebRoot and provide as static files.

The build binaries should be generated on src/Backend/bin/Debug/netcoreapp1.1/ directory.

Unlicensed

Manoel Vilela

• Git
• Hugo

We use --recursive (from most code editors, “Git Clone (Recursive)” as opposed to “Git Clone”) to ensure the theme subdmodule is cloned, too.

git clone --recursive git@github.com:phi-math-website/phi-math-website.git

If you leave this command running and modify content, your local test website instance will automatically update.

hugo server -O

If you make changes to some configuration values, the instance might not update. In this case. Stop and rerun the command instead.

Once you are satisfied with your modifications, run:

hugo

This will generate your website in a new public/ folder. Next, we’ll see how to upload this to events.illc.uva.nl/Phi-Math.

• SSH, or any program like SFTP, SCP or RSync that can work over the SSH protocol.

You can email webmaster Marco Vervoort your UvAnetID (e.g., 74958495) explaining that you are a new $\Phi$-Math board member. Marco will grant you access. For the remainder of this guide, we will refer to your UvAnetID as UVANETID within commands/configuration files.

Logging in directly to the server with SSH is blocked by ICTS. Instead you log in via a ‘gateway server’ with hostname pascal.ic.uva.nl. For testing you can log in with SSH using ssh UVANETID@pascal.ic.uva.nl and then login by executing the command on this server ssh UVANETID@goedel.fnwi.uva.nl.

For practical use you can automate this ‘log through’. To do this:

1. create a .ssh folder in your home/user directory (e.g., /home/marco or C:\Users\marco)
2. inside the newly created folder, create a config file (with no file extension) that looks like this:

Host phimathproxy
  Hostname pascal.ic.uva.nl
  User UVANETID

Host phimath
  Hostname goedel.fnwi.uva.nl
  ProxyJump phimathproxy
  User UVANETID

This ensures that when you run the command ssh phimath on your own computer, SSH will first log onto the gateway server and then do the ‘log through’ automatically. And this will also configure other programs based on SSH (like the aforementioned SFTP, SCP and Rsync) to use the same mechanism to access goedel.fnwi.uva.nl.

Initially, this setup means that you have to enter your UvANetID password twice. If you log in separately with SSH on pascal.ic.uva.nl(using ssh UVANETID@pascal.ic.uva.nl) you can use the command vi_authorized_keys to add a public SSH key (using the vi text editor from the terminal), so that you no longer have to enter a password for the pascal.ic.uva.nl login. If you don’t have a public SSH key yet, you can create one by running the command on your own computer (not the web server) ssh-keygen then two files will be created on your own computer in the subdirectory ‘.ssh’, ‘.ssh/id_rsa.pub’ (the public SSH key whose content you are using above) and ‘.ssh/id_rsa’ (a private SSH key that you should not share with others and that SSH uses when logging in).

On Linux systems, if you automate the ‘log-through’ for SSH as described above, the other programs based on SSH (like the aforementioned SFTP, SCP and Rsync) will also use this configuration to access the server. So you can just run things like:

rsync -r public/* phimath:/var/www/illc/events/Phi-Math/

or (since not all windows machines come with the rsync command preinstalled)

scp -r public/* phimath:/var/www/illc/events/Phi-Math/

After logging in, the site files may be found in the filesystem at /var/www/illc/events/Phi-Math. You can modify them there.

Finally, Marco has a request that he would ask us to keep in mind when managing the site files: can we make sure that all files and subdirectories are created with ‘group-writeable’ permissions, and all subdirectories also with the ‘set-group-id’ flag? This ensures that files and directories belong to the user group www-illc-phimath-science and are editable by all members of this group. To do it manually, we can use the commands

chmod -R g+w /var/www/illc/events/Phi-Math/*
chgrp -R www-illc-phimath-science /var/www/illc/events/Phi-Math/*
find /var/www/illc/events/Phi-Math/* -type d -exec chmod g+s {} +

(Make sure to press enter after pasting this in your terminal!)

If we forgot this step, the other $\Phi$-Math board member will be unable to edit these files!

For general hosting Marco has a few more requests, but those are not relevant for $\Phi$-Math, as we are using the ‘create static site on your own computer and then upload the pages’ approach. If we ever want to switch, we should contact Marco first.

Accelerated NLP pipelines for fast inference 🚀 on CPU. Built with 🤗Transformers and ONNX runtime.

pip install git+https://github.com/patil-suraj/onnx_transformers

NOTE : This is an experimental project and only tested with PyTorch

The pipeline API is similar to transformers pipeline with just a few differences which are explained below.

Just provide the path/url to the model and it’ll download the model if needed from the hub and automatically create onnx graph and run inference.

from onnx_transformers import pipeline

# Initialize a pipeline by passing the task name and 
# set onnx to True (default value is also True)
>>> nlp = pipeline("sentiment-analysis", onnx=True)
>>> nlp("Transformers and onnx runtime is an awesome combo!")
[{'label': 'POSITIVE', 'score': 0.999721109867096}]  

Or provide a different model using the model argument.

from onnx_transformers import pipeline

>>> nlp = pipeline("question-answering", model="deepset/roberta-base-squad2", onnx=True)
>>> nlp({
  "question": "What is ONNX Runtime ?", 
  "context": "ONNX Runtime is a highly performant single inference engine for multiple platforms and hardware"
})
{'answer': 'highly performant single inference engine for multiple platforms and hardware', 'end': 94, 'score': 0.751201868057251, 'start': 18}

Set onnx to False for standard torch inference.

You can create Pipeline objects for the following down-stream tasks:

• feature-extraction: Generates a tensor representation for the input sequence
• ner: Generates named entity mapping for each word in the input sequence.
• sentiment-analysis: Gives the polarity (positive / negative) of the whole input sequence. Can be used for any text classification model.
• question-answering: Provided some context and a question referring to the context, it will extract the answer to the question in the context.
• zero-shot-classification:

Calling the pipeline for the first time loads the model, creates the onnx graph, and caches it for future use. Due to this, the first load will take some time. Subsequent calls to the same model will load the onnx graph automatically from the cache.

The key difference between HF pipeline and onnx_transformers is that the model parameter should always be a string (path or url to the saved model). Also, the zero-shot-classification pipeline here uses roberta-large-mnli as default model instead of facebook/bart-large-mnli as BART is not yet tested with onnx runtime.

Note: For some reason, onnx is slow on colab notebook so you won’t notice any speed-up there. Benchmark it on your own hardware.

For detailed benchmarks and other information refer to this blog post and notebook.

• Accelerate your NLP pipelines using Hugging Face Transformers and ONNX Runtime
• Exporting 🤗 transformers model to ONNX

To benchmark the pipelines in this repo, see the benchmark_pipelines notebook.

(Note: These are not yet comprehensive benchmarks.)

Benchmark feature-extraction pipeline

Benchmark question-answering pipeline

>As the current housing systems are moving towards automation, the focus on the systems used within the house is given more focus than the customer requirement. The systems available in the current market are complex and expensive. The objective of the “IoT based remote sensor data monitoring and actuator control” project is to create a partial open-source monitoring system that can be customized based on the individual requirements of the customer which is cheaper than the available market alternatives and user-friendly.

Since this monitoring system can be used for various applications, it is difficult to reproduce every possible scenario. Hence the prototype for the use case of home water tank was built.

DHT11 - temperature and humidity readings.

water level sensor - To avoid overflow of the water tank and emergency stop when desired level is reached

MQTT - Use as communication protocol with between  sensor, between sensor and relay

Raspberry pi zero - For getting the sensor data
Raspberry pi 4 - To control relay & power supply

HiveMQ broker - It was used to test published/subscribed messages.

Influx-DB - It was used as Database For storing water level and DHT11 sensor streaming time-series data.

Grafana - It was used for visualizing temperature and humidity streaming time-series data

The above picture illustrates the monitoring system of the water level in a smart home network. Water level sensor and DHT11 sensor are connected to Raspberry pi zero.Firstly, Raspberry-pi Zero is gathered streaming data from DHT11 – sensor, the water level – sensor and the ultrasonic sensor. Which values are converted to the action or messages and Those messages are published to the MQTT – Protocol to a particular topic. Next, RaspberryPi- 4 is giving a request to the MQTT protocol in order to get sensor values that are published by Raspberrypi Zero. Moreover, According to the message relay is handled by RaspberryPi- 4 in order to manage the power supply for the motor as well as Raspberry pi- 4 push streaming time-series data and message into the InfluxDB. After that, Grafana is connected with InfluxDB to do streaming time-series data visualization.

react-redux-modal-provider controls the state of your React modal components using Redux.

npm i --save react-redux-modal-provider

import ModalProvider from 'react-redux-modal-provider';

export default render(
  <Provider store={store}>
    <div>
      <App />
      <ModalProvider />
    </div>
  </Provider>,
  document.getElementById('app')
);

import { reducer as modalProvider } from 'react-redux-modal-provider';

export default combineReducers({
  modalProvider,
});

// app.jsx
import { showModal } from 'react-redux-modal-provider';
import MyModal from './myModal';

export default (props) => (
  <div>
    <p>
      Hello world
    </p>
    <button
      type="button"
      onClick={() => showModal(MyModal, { message: 'Hello' })}>
      Present modal
    </button>
  </div>
);

// myModal.jsx
import { Modal } from 'react-bootstrap';

export default (props) => (
  <Modal show={props.show}>
    <Modal.Body>
      {props.message}
    </Modal.Body>

    <Modal.Footer>
      <Button onClick={props.hideModal}>Ok</Button>
    </Modal.Footer>
  </Modal>
);

show and hideModal props are passed in automatically.

This is the default ModalProvider implementation. Each new modal stacks up on top of previous one.

import { StackableModalProvider } from 'react-redux-modal-provider';

export default render(
  <Provider store={store}>
    <div>
      <App />
      <StackableModalProvider />
    </div>
  </Provider>,
  document.getElementById('app')
);

One modal at a time. Each new modal triggers hideModal on previous one.

import { SingleModalProvider } from 'react-redux-modal-provider';

export default render(
  <Provider store={store}>
    <div>
      <App />
      <SingleModalProvider />
    </div>
  </Provider>,
  document.getElementById('app')
);

1. You don’t have to think about where your modal component should fit into component tree, because it doesn’t really matter where to render a modal.

2. No need to connect() your modal component to Redux, unless you want it to be able to create other modals itself.

• Thanks @yesmeck, author of redux-modal, for webpack config I borrowed.

• Thanks @diegoddox, author of react-redux-toastr, for the idea how to dispatch actions from anywhere using EventEmitter.

MIT

A single-class pure VB6 library for zip archives management

Just include cZipArchive.cls to your project and start using instances of the class like this:

With New cZipArchive
    .AddFile App.Path & "\your_file"
    .CompressArchive App.Path & "\test.zip"
End With

With New cZipArchive
    .AddFromFolder "C:\Path\To\*.*", Recursive:=True
    .CompressArchive App.Path & "\archive.zip"
End With

With New cZipArchive
    .OpenArchive App.Path & "\test.zip"
    .Extract "C:\Path\To\extract_folder"
End With

Method Extract can optionally filter on file mask (e.g. Filter:="*.doc"), file index (e.g. Filter:=15) or array of booleans with each entry to decompress index set to True.

OutputTarget can include a target new_filename to be used when extracting a specific file from the archive.

With New cZipArchive
    .OpenArchive App.Path & "\test.zip"
    .Extract "C:\Path\To\extract_folder\new_filename", Filter:="your_file"
End With

By using FileInfo property keyed on entry filename in first parameter and zipIdxSize like this

With New cZipArchive
    .OpenArchive App.Path & "\test.zip"
    Debug.Print .FileInfo("report.pdf", zipIdxSize)
End With

By using FileInfo propery keyed on entry numeric index in first parameter like this

Dim lIdx            As Long
With New cZipArchive
    .OpenArchive App.Path & "\test.zip"
    For lIdx = 0 To .FileCount - 1
        Debug.Print "FileName=" & .FileInfo(lIdx, zipIdxFileName) & ", Size=" & .FileInfo(lIdx, zipIdxSize)
    Next
End With

Here is a list of available values for the second parameter of FileInfo:

Make sure to set Conditional Compilation in Make tab in project’s properties dialog to include ZIP_CRYPTO = 1 setting for crypto support to get compiled from sources. By default crypto support is not compiled to reduce footprint on the final executable size.

With New cZipArchive
    .OpenArchive App.Path & "\test.zip"
    .Extract App.Path & "\test", Password:="123456"
End With

Use Password parameter on AddFile method together with EncrStrength parameter to set crypto used when creating archive.

Note that default ZipCrypto encryption is weak but this is the only option which is compatible with Windows Explorer built-in zipfolders support.

Sample utility function ReadBinaryFile in /test/basic/Form1.frm returns byte array with file’s content.

Dim baZip() As Byte
With New cZipArchive
    .AddFile ReadBinaryFile("sample.pdf"), "report.pdf"
    .CompressArchive baZip
End With
WriteBinaryFile "test.zip", baZip

Method Extract accepts byte array target too.

Dim baOutput() As Byte
With New cZipArchive
    .OpenArchive ReadBinaryFile("test.zip")
    .Extract baOutput, Filter:=0    '--- archive's first file only
End With

- Deflate64 (de)compressor
- VBA7 (x64) support

1. Скачать и установить контейнер Java-сервлетов(например, Apache Tomcat)
2. Подготовить базу данных для приложения
   • Создать базу данных PostgreSQL
   • Выполнить скрипты из /databaseScripts/databaseScripts.sql в новой базе данных(для создания схемы)
3. Создать war-артифакт приложения
   • Выполнить команду ./gradlew war
   • Готовый артифакт приложения появится в директории /build/libs/
4. Развернуть war-файл в контейнере сервлетов
5. Запросы к API можно отправлять через URL вида: http://localhost:8080/meetup-rest-api-0-0-1/meetup

• Получить все митапы

  Пример запроса:

GET http://localhost:8080/meetup-rest-api-0-0-1/meetup HTTP/1.1
Content-Type: application/json

{
"filterParameters": {
"agenda": "meet"
},
"sortingParameters":[
"agenda"
]
}

где filterParameters и sortingParameters являются опциональными параметрами, задающими фильтрацию и сортировку
результатов. Поддерживаемые параметры митапов: agenda, dateTime, organizer

• Получить митап

  Пример запроса:

GET http://localhost:8080/meetup-rest-api-0-0-1/meetup/{id} HTTP/1.1

где {id} – id запрашиваемого митапа

• Добавить новый митап

  Пример запроса:

PUT http://localhost:8080/meetup-rest-api-0-0-1/meetup HTTP/1.1
Content-Type: application/json

{
"agenda": "Very Important Things",
"description": "We discuss some very important Things",
"organizer": "An important person",
"dateTime": "17.11.2022 11:41",
"location": "An important office"
}

• Удалить митап

  Пример запроса:

DELETE http://localhost:8080/meetup-rest-api-0-0-1/meetup/{id} HTTP/1.1

где {id} – id удаляемого митапа

• Изменить митап

  Пример запроса:

POST http://localhost:8080/meetup-rest-api-0-0-1/meetup/{id} HTTP/1.1
Content-Type: application/json

{
"agenda": "Very Important Things",
"description": "We discuss some very important Things",
"organizer": "An important person",
"dateTime": "17.11.2022 11:41",
"location": "An important office"
}

где {id} – id изменяемого митапа

• James Dinan (MPI-2 implementation)
• Jeff Hammond (MPI-3 implementation)

This project provides a full, high performance, portable implementation of the ARMCI runtime system using MPI’s remote memory access (RMA) functionality.

See Travis for failure details. All recent failures have been caused by dependencies (system toolchain or MPI library).

ARMCI-MPI uses autoconf and must be configured before compiling:

 $ ./configure

Many configure options are provided, run configure --help for details. After configuring the source tree, the code can be built and installed by running:

 $ make && make install

The quality of MPI-RMA implementations varies. We recommend that you always use the absolute latest release or release candidate version of MPI unless you are aware of a specific issues that prevents this.

With the exception of the IBM Blue Gene platforms, all MPI libraries known to the ARMCI-MPI developers support MPI-3 RMA, hence the MPI-2 RMA implementation has been deprecated and is no longer supported. If you are using ARMCI-MPI on a Blue Gene system, use the legacy branch or contact the developers for assistance.

As of September 2018, MPICH 3.3b3 and Open-MPI 3.1.2 are passing all of the tests in Travis CI. Information about other implementations will be added here soon.

As of April 2014, the following implementations were known to work correctly with ARMCI-MPI (MPI-3 version):

• MPICH 3.0.4 and later on Mac, Linux SMPs and SGI SMPs.
• MVAPICH2 2.0a and later on Linux InfiniBand clusters.
• CrayMPI 6.1.0 and later on Cray XC30.
• SGI MPT 2.09 on SGI SMPs.
• Open-MPI development version on Mac (set ARMCI_STRIDED_METHOD=IOV and ARMCI_IOV_METHOD=BATCHED)

Note that a bug in MPICH 3.0 or 3.1 that propagated to MVAPICH2, Cray MPI and Intel MPI affects correctness when windows are backed by shared-memory. This bug affects ARMCI_Rmw and is avoided by setting ARMCI_USE_WIN_ALLOCATE=0 in your runtime environment. This may negatively affect performance in some cases and prevents one from using Casper, hence is not the default.

As of August, 2011 the following MPI-2 implementations were known to work correctly with ARMCI-MPI (MPI-2 version):

• MPICH2 and MPICH 3+
• MVAPICH2 1.6
• Cray MPI on Cray XE6
• IBM MPI on BG/P (set ARMCI_STRIDED_METHOD=IOV and ARMCI_IOV_METHOD=BATCHED for performance reasons)
• Open-MPI 1.5.4 (set ARMCI_STRIDED_METHOD=IOV and ARMCI_IOV_METHOD=BATCHED for correctness reasons)

The following MPI-2 implementations are known to fail with ARMCI-MPI:

• MVAPICH2 prior to 1.6

To build GA (version 5.2 or later) with ARMCI-MPI (any version), use the configure option --with-armci=$(PATH_TO_ARMCI_MPI) and make sure that you use the same MPI implementation with GA that was used to compile ARMCI-MPI.

ARMCI-MPI (MPI-3) has been tested extensively with GA since version 5.2.

If you are an NWChem user, you can use ${NWCHEM_TOP}/src/tools/install-armci-mpi without having to download or build ARMCI-MPI manually.

ARMCI-MPI includes a set of testing and benchmark programs located under tests/ and benchmarks/. These programs can be compiled and run via:

$ make check MPIEXEC="mpiexec -n 4"

The MPIEXEC variable is optional and is used to override the default MPI launch command. If you want only to build the test suite, the following target can be used:

$ make checkprogs

• Because of MPI-2’s semantics, you are not allowed to access shared memory directly, it must be through put/get. Alternatively you can use the new ARMCI_Access_begin/end() functions.

• MPI-3 allows direct access provided one uses a synchronization operation afterwards. The ARMCI_Access_begin/end() functions are also valid.

• On some MPI implementations and networks you may need to enable implicit progress. In many cases this is done through an environment variable. For MPICH2: set MPICH_ASYNC_PROGRESS; for MVAPICH2 recompile with --enable-async-progress and set MPICH_ASYNC_PROGRESS; set DCMF_INTERRUPTS=1 for MPI on BGP; etc.

See this page for more information on activating asynchronous progress in MPI. However, we find that most platforms show no improvement and often a decrease in performance, provided the application makes calls to GA/ARMCI/MPI frequently enough on all processes.

We recommend the user of Casper for asynchronous progress in ARMCI-MPI. See the Casper website for details.

Boolean environment variables are enabled when set to a value beginning with ‘t’, ‘T’, ‘y’, ‘Y’, or ‘1’; any other value is interpreted as false.

ARMCI_VERBOSE (boolean)

Enable extra status output from ARMCI-MPI.

ARMCI_DEBUG_ALLOC (boolean)

Turn on extra shared allocation debugging.

ARMCI_FLUSH_BARRIERS (boolean) (deprecated)

Enable/disable extra communication flushing in ARMCI_Barrier. Extra flushes are present to help make unsafe DLA safer. (This option is deprecated with the ARMCI-MPI3 implementation.)

ARMCI_CACHE_RANK_TRANSLATION (boolean)

Create a table to more quickly translate between absolute and group ranks.

ARMCI_PROGRESS_THREAD (boolean)

Create a Pthread to poke the MPI progress engine.

ARMCI_PROGRESS_USLEEP (int)

Argument to usleep() to pause the progress polling loop.

ARMCI_NONCOLLECTIVE_GROUPS (boolean)

Enable noncollective ARMCI group formation; group creation is collective on the output group rather than the parent group.

ARMCI_SHR_BUF_METHOD = { COPY (default), NOGUARD }

ARMCI policy for managing shared origin buffers in communication operations: lock the buffer (unsafe, but fast), copy the buffer (safe), or don’t guard the buffer – assume that the system is cache coherent and MPI supports unlocked load/store.

ARMCI_STRIDED_METHOD = { DIRECT (default), IOV }

Select the method for processing strided operations.

ARMCI_IOV_METHOD = { AUTO (default), CONSRV, BATCHED, DIRECT }

Select the IO vector communication strategy: automatic; a “conservative” implementation that does lock/unlock around each operation; an implementation that issues batches of operations within a single lock/unlock epoch; and a direct implementation that generates datatypes for the origin and target and issues a single operation using them.

ARMCI_IOV_CHECKS (boolean)

Enable (expensive) IOV safety/debugging checks (not recommended for performance runs).

ARMCI_IOV_BATCHED_LIMIT = { 0 (default), 1, … }

Set the maximum number of one-sided operations per epoch for the BATCHED IOV method. Zero (default) is unlimited.

Return a typed array constructor for creating typed arrays stored in little-endian byte order.

npm install @stdlib/array-little-endian-factory

var littleEndianFactory = require( '@stdlib/array-little-endian-factory' );

var Float64ArrayLE = littleEndianFactory( 'float64' );
// returns <Function>

var Float32ArrayLE = littleEndianFactory( 'float32' );
// returns <Function>

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE();
// returns <Float64ArrayLE>

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( 5 );
// returns <Float64ArrayLE>

var Float32Array = require( '@stdlib/array-float32' );

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr1 = new Float32Array( [ 0.5, 0.5, 0.5 ] );
var arr2 = new Float64ArrayLE( arr1 );
// returns <Float64ArrayLE>

var v = arr2.get( 0 );
// returns 0.5

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( [ 0.5, 0.5, 0.5 ] );
// returns <Float64ArrayLE>

var v = arr.get( 0 );
// returns 0.5

var ArrayBuffer = require( '@stdlib/array-buffer' );

var Float64ArrayLE = littleEndianFactory( 'float64' );

var buf = new ArrayBuffer( 32 );
var arr = new Float64ArrayLE( buf, 0, 4 );
// returns <Float64ArrayLE>

var Float64ArrayLE = littleEndianFactory( 'float64' );

var nbytes = Float64ArrayLE.BYTES_PER_ELEMENT;
// returns 8

var Float64ArrayLE = littleEndianFactory( 'float64' );

var str = Float64ArrayLE.name;
// returns 'Float64ArrayLE'

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( 5 );
var buf = arr.buffer;
// returns <ArrayBuffer>

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( 5 );
var byteLength = arr.byteLength;
// returns 40

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( 5 );
var byteOffset = arr.byteOffset;
// returns 0

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( 5 );
var nbytes = arr.BYTES_PER_ELEMENT;
// returns 8

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( 5 );
var len = arr.length;
// returns 5

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = Float64ArrayLE.from( [ 1.0, -1.0 ] );
// returns <Float64ArrayLE>

var v = arr.get( 0 );
// returns 1.0

function mapFcn( v ) {
    return v * 2.0;
}

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = Float64ArrayLE.from( [ 1.0, -1.0 ], mapFcn );
// returns <Float64ArrayLE>

var v = arr.get( 0 );
// returns 2.0

function mapFcn( v ) {
    this.count += 1;
    return v * 2.0;
}

var Float64ArrayLE = littleEndianFactory( 'float64' );

var ctx = {
    'count': 0
};

var arr = Float64ArrayLE.from( [ 1.0, -1.0 ], mapFcn, ctx );
// returns <Float64ArrayLE>

var v = arr.get( 0 );
// returns 2.0

var n = ctx.count;
// returns 2

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = Float64ArrayLE.of( 1.0, -1.0 );
// returns <Float64ArrayLE>

var v = arr.get( 0 );
// returns 1.0

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( 10 );

// Set the first element:
arr.set( 1.0, 0 );

// Get the first element:
var v = arr.get( 0 );
// returns 1.0

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( 10 );

var v = arr.get( 100 );
// returns undefined

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( [ 1.0, 2.0, 3.0 ] );
// returns <Float64ArrayLE>

var v = arr.get( 0 );
// returns 1.0

v = arr.get( 1 );
// returns 2.0

// Set the first two array elements:
arr.set( [ 4.0, 5.0 ] );

v = arr.get( 0 );
// returns 4.0

v = arr.get( 1 );
// returns 5.0

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( [ 1.0, 2.0, 3.0 ] );
// returns <Float64ArrayLE>

// Set the last two array elements:
arr.set( [ 4.0, 5.0 ], 1 );

var v = arr.get( 1 );
// returns 4.0

v = arr.get( 2 );
// returns 5.0

var Float64ArrayLE = littleEndianFactory( 'float64' );

var arr = new Float64ArrayLE( [ 1.0, 2.0, 3.0 ] );

var str = arr.toString();
// returns '1,2,3'

var Float64Array = require( '@stdlib/array-float64' );
var logEach = require( '@stdlib/console-log-each' );
var littleEndianFactory = require( '@stdlib/array-little-endian-factory' );

var Float64ArrayLE = littleEndianFactory( 'float64' );

// Create a typed array by specifying a length:
var out = new Float64ArrayLE( 3 );
logEach( '%s', out );

// Create a typed array from an array:
var arr = [ 1.0, -1.0, -3.14, 3.14, 0.5, 0.5 ];
out = new Float64ArrayLE( arr );
logEach( '%s', out );

// Create a typed array from an array buffer:
arr = new Float64Array( [ 1.0, -1.0, -3.14, 3.14, 0.5, 0.5 ] ); // host byte order
out = new Float64ArrayLE( arr.buffer );
logEach( '%s', out );

// Create a typed array from an array buffer view:
arr = new Float64Array( [ 1.0, -1.0, -3.14, 3.14, 0.5, 0.5 ] ); // host byte order
out = new Float64ArrayLE( arr.buffer, 8, 2 );
logEach( '%s', out );