Call a subroutine on destruction of a variable.

Changes for 0.09 - 2024-03-29T19:04:01+01:00

• Further fix on older perls

Tied objects

Changes for 0.16 - 2024-03-29

• Fix before, around and after to only work on readonly functions (properties that have a value as a code ref), to prevent unexpected behaviour.

Tied objects

Changes for 0.15 - 2024-03-29

• Adds get_initialised function which can be called on Rope to retrieve an already initialised object by index.
• Adds before, around and after keywords which can be chained and work for all properties/functions, not traditional subs currently because they are effectively 'hidden' to Rope.

Hi all,

I have most difficulties to make my LWP::UserAgent to use TLS 1.2/1.3

I tried :

my $ua = new LWP::UserAgent( 'ssl_opts' => { SSL_version => 'TLSv12:!SSLv2:!SSLv3:!TLSv1:!TLSv11', } );

but no success .....

also tried :

$ENV{https_version}=3;

perl v5.32 is not that old, I am very surprised I can't establish connecting to TLS 1.2/1.3 server

Is it a known problem with openSSL version used to compile v5.32 ? if so, can I upgrade the SSL libs only ? or is there an option to pass on to force TLS 1.2/1.3 ?

Windows 10 - strawberry v5.32.1 MSWin32-x64-multi-thread

Thank you very much !

​

submitted by /u/lowpowerdesign
[link] [comments]

Read environment variables from .env file

Changes for 0.012 - 2024-03-29T18:05:10+02:00

• Fix lib and test code to use Cwd::abs_path with File::Spec->catdir|catfile.

Class for conversion between percent number to star visualization

Changes for 0.03 - 2024-03-29T16:01:21+01:00

• Add missing dot in doc.
• Improve doc for constructor.
• Move bugtracker to Github.
• Move valid tests to top.
• Rewrite check of error to Mo::utils.

I have a bunch of YAML config files (using YAML::XS) which are multidimensional, and the values can be HASH, ARRAY, or SCALAR. Right now, I have a kind of a template YAML that defines the layout that someone can use as a reference to write a config file. The keys and layout are the same, except instead of having values it has two keys: 1) required (boolean) and 2) type (ARRAY, HASH, STR, etc). What I would like is for a sub to just verify the tree and spit out an error on which key is the problem, but I'm getting kind of lost on trying to loop both trees together.

Here is a small example from a part of a YAML config.

split: level: 3 filter: directories: delete: - \/main.assets$ keep: - S2589004224001688 files: keep: - issue.xml - main.xml - main.pdf 

split: required: True type: HASH group: required: False type: HASH by: required: True type: STR count: required: True type: STR level: required: True type: INT filter: required: False type: HASH directories: required: False type: HASH delete: required: False type: ARRAY keep: required: False type: ARRAY files: required: False type: HASH delete: required: False type: ARRAY keep: required: False type: ARRAY 

submitted by /u/sirhalos
[link] [comments]

Some fun with chatgpt (note that you can get the same information with a rudimentary search across various research databases followed by statistical analysis of the tags)

When reading the chat bot response note that

legacy systems include many/most of the research tools in many fields many of the legacy systems are heavily dependent on bash for gluing their individual components many of these legacy systems include a noninsignificant amount for web programming Python / R etc just glue over code written in C and Fortran, ie they just provide an API

Prompt: can you suggest a few application areas for perl as glue in scientific component based software engineering?

Perl can be effectively used as "glue" in scientific component-based software engineering due to its strong text manipulation capabilities, mature ecosystem, and ability to interface with various systems and languages. Here are a few application areas where Perl can excel as a glue language:

1. Bioinformatics Pipelines

Perl has been traditionally strong in bioinformatics, where it is used to connect different software components for tasks like DNA sequence analysis, protein structure prediction, and genetic data processing. It can orchestrate workflows that involve calling various tools, parsing their output, and preparing it for the next step in the analysis pipeline.

2. Data Transformation and Integration

In scientific computing, there is often a need to transform data between different formats or integrate data from various sources. Perl can serve as a glue language to script these transformations and integrations, handling formats like CSV, JSON, XML, and proprietary binary formats commonly used in scientific instruments.

3. System Administration for Research Computing

Perl can automate and manage tasks in research computing environments, such as scheduling jobs on a compute cluster, monitoring resource usage, and managing file systems. It can glue together various system components, ensuring they work seamlessly to support scientific computation.

4. Legacy Scientific Software Integration

Many scientific research fields have legacy software components written in Perl or that provide Perl interfaces. Perl can be used to integrate these components with newer systems, preserving the value of previous investments in software development.

5. Automated Reporting and Visualization

Perl can be used to generate automated reports and visualizations from scientific data. By gluing together data extraction, analysis, and reporting tools, Perl scripts can produce regular, up-to-date reports or visualizations of ongoing scientific research or experiments.

Limitations to Consider

While Perl is useful in these areas, it's essential to consider its limitations. The ecosystem around Perl, particularly in scientific computing, may not be as vibrant as it once was, with many new developments occurring in other languages like Python. Therefore, while Perl remains a powerful tool for certain tasks, its role in new scientific software projects might be more limited compared to its peak usage times.

In conclusion, Perl's role as glue in scientific component-based software engineering is well established, particularly in areas where text processing and legacy system integration are paramount. However, the choice to use Perl should be balanced with considerations about the current and future needs of the project, as well as the availability of skills and community support.

submitted by /u/ReplacementSlight413
[link] [comments]