DEV Community: Lucas Cimon

fpdf2 latest news

Lucas Cimon — Mon, 14 Oct 2024 15:47:50 +0000

You can find a selection of the features & improvements in fpdf2 made over the last 18 months in this blog post: https://chezsoi.org/lucas/blog/fpdf2-latest-news.html

Setting up linters in Gitlab CI for C++ and Groovy / Jenkins code

Lucas Cimon — Fri, 03 May 2024 11:42:05 +0000

A very short blog post to share some minimal code snippets on how to quickly and easily setup static code analysis tools on C++ code and Jenkins pipelines (or any Groovy code), for example in Gitlab CI pipelines.

Linting C++ code with clang-tidy

clang-tidy is a clang-based C++ linter tool that can identify and auto-fix some programming errors, like style violations, interface misuse, or bugs that can be deduced via static analysis.



cpp-linter:
  # No need to specify the next line if your default Docker image is already an alpine:
  image: alpine:3.17
  script:
    - apk add clang-extra-tools
    - find . -name '*.cpp' -exec clang-tidy --fix {} \; | tee clang-tidy.log
    # Fail the job if errors were found:
    - ! grep "Found compiler error" clang-tidy.log
  rules:
    # Executing for every commit on the main branch where C++ files were modified:
    - if: $CI_PIPELINE_SOURCE == "push" && $CI_COMMIT_BRANCH == "main"
      changes:
        - "*.cpp"
        - "**/*.cpp"
    # Executing for every commit in a non-draft merge request where C++ files were modified
    - if: $CI_PIPELINE_SOURCE == "merge_request_event" && $CI_MERGE_REQUEST_TITLE !~ /^Draft:.*$/
      changes:
        - "*.cpp"
        - "**/*.cpp"

You should also initiate a .clang-tidy configuration file like this:



Checks: '*,-llvmlibc-*'
WarningsAsErrors: '*'

This default file is a strict starting point: it enables all rules, except the ones specific to LLVM, and treat all warnings as errors.
You can then add -${ruleName} to the Checks entry in this file to disable some checks.
Check the clang-tidy documentation for more details about rules and suppressing errors & warnings using code comments.

You can also generate an exhaustive .clang-tidy configuration file, with an extra CheckOptions field listing all default values for rules parameters, by running this command:



clang-tidy -checks='*,-llvmlibc-*' -warnings-as-errors='*' -dump-config > .clang-tidy

Linting Groovy / Jenkins code with CodeNarc

CodeNarc analyzes Groovy code for defects, bad practices, inconsistencies, style issues and more.

It can be used to perform static code analysis on Jenkins pipelines.

Because executing CodeNarc from the command-line is not so simple, I find it easier to use Gradle and its dedicated plugin to execute CodeNarc:



groovy-linter:
  image: gradle:8.7
  script:
    - gradle check
  rules:
    # Executing for every commit on the main branch where C++ files were modified:
    - if: $CI_PIPELINE_SOURCE == "push" && $CI_COMMIT_BRANCH == "main"
      changes:
        - "*.groovy"
        - "**/*.groovy"
    # Executing for every commit in a non-draft merge request where Groovy files were modified
    - if: $CI_PIPELINE_SOURCE == "merge_request_event" && $CI_MERGE_REQUEST_TITLE !~ /^Draft:.*$/
      changes:
        - "*.groovy"
        - "**/*.groovy"

You will also need a build.gradle file in the directory where the gradle command is executed:



plugins {
    id 'groovy'
    id 'codenarc'
}
repositories {
    jcenter()
    // Required if you have Jenkins dependencies:
    maven {
        url('https://repo.jenkins-ci.org/public/')
    }
}
def codeNarcVersion = '3.4.0'
dependencies {
    // CodeNarc must be added there, or you will get a ClassNotFoundException:
    codenarc group: 'org.codenarc', name: 'CodeNarc', version: codeNarcVersion
    codenarc group: 'org.jenkins-ci.plugins', name: 'plugin', version: '4.79'
}
sourceSets {
    main {
        groovy {
            // Specify the relative paths to the directories containing your .groovy files:
            srcDirs = ['.']
        }
    }
}
codenarc {
    configFile = file('./codenarc_rules.groovy')
    toolVersion = codeNarcVersion
    reportFormat = 'text'
}
// The two following blocks allow to display the CodeNarc report in the console:
// ( this recipe comes from this SO answer: https://stackoverflow.com/a/36899862/636849 )
task codenarcConsoleReport {
    doLast {
        println file("${codenarc.reportsDir}/main.txt").text
    }
}
codenarcMain {
    finalizedBy codenarcConsoleReport
    compilationClasspath = sourceSets.main.compileClasspath + sourceSets.main.output //+ files('../relative/path/to/shared/lib/dir')
}
// You can optionnally disable the Groovy compilation step, to execute faster:
compileGroovy.enabled = false

And finally you can initialize a configuration file for the linter, named codenarc_rules.groovy:

codenarc_rules.groovy

ruleset {

    ruleset('rulesets/basic.xml')

    ruleset('rulesets/braces.xml')

    ruleset('rulesets/comments.xml')

    ruleset('rulesets/concurrency.xml')

    ruleset('rulesets/convention.xml') {

        // You may want to disable some rules, like this:

        CompileStatic(enabled:false)

        MethodParameterTypeRequired(enabled:false)

        MethodReturnTypeRequired(enabled:false)

        NoDef(enabled:false)

        PublicMethodsBeforeNonPublicMethods(enabled:false)

        VariableTypeRequired(enabled:false)

        TrailingComma(enabled:false)

    }

    ruleset('rulesets/design.xml')

    ruleset('rulesets/dry.xml') {

        // Allowing several benign cases of code duplication:

        DuplicateListLiteral(enabled:false)

        DuplicateMapLiteral(enabled:false)

        DuplicateNumberLiteral(enabled:false)

        DuplicateStringLiteral(enabled:false)

    }

    ruleset('rulesets/enhanced.xml')

    ruleset('rulesets/exceptions.xml')

    ruleset('rulesets/formatting.xml') {

        // Disabling rules that are too strict for my project:

        BlockEndsWithBlankLine(enabled:false)

        BlockStartsWithBlankLine(enabled:false)

        ConsecutiveBlankLines(enabled:false)

        Indentation(enabled:false)

        LineLength(enabled:false)

        SpaceAfterOpeningBrace(enabled:false)

        SpaceAroundMapEntryColon(enabled:false)

        SpaceBeforeClosingBrace(enabled:false)

    }

    ruleset('rulesets/generic.xml')

    // ruleset('rulesets/grails.xml')

    ruleset('rulesets/groovyism.xml')

    ruleset('rulesets/imports.xml')

    // ruleset('rulesets/jdbc.xml')

    // ruleset('rulesets/junit.xml')

    ruleset('rulesets/logging.xml') {

        // Allowing to print to stderr & stdout

        SystemErrPrint(enabled:false)

        SystemOutPrint(enabled:false)

    }

    ruleset('rulesets/naming.xml')

    ruleset('rulesets/security.xml')

    ruleset('rulesets/serialization.xml')

    ruleset('rulesets/size.xml')

    ruleset('rulesets/unnecessary.xml')

    ruleset('rulesets/unused.xml')

}

The details for all rules can be found there: codenarc.org/codenarc-rule-index.html

Original post on my blog: https://chezsoi.org/lucas/blog/setting-up-linters-in-gitlab-ci-for-c-and-groovy-jenkins-code.html

Pylint strict base configuration

Lucas Cimon — Wed, 03 May 2023 11:31:51 +0000

Pylint is a great static code analyser for Python code.
I have been using it for several years, in various projects, and it's simple to use yet very powerful.

I even contributed to Pylint by submitting a new rule a few years ago : implicit-str-concat.

For an introduction to Pylint, you can check those tutorials:

official documentation: Tutorial & Usage
How To Get Started With Pylint by Oliyadk @medium.com

Pylint default configuration is very reasonable, but one thing that I find missing is a simple way to switch to a "strict" mode, that would enable all optional rules. ESLint has eslint-config-strict-mode for example.

Such "strict" mode can be useful:

when bootstrapping a new project, you may want to enable all rules by default,
and progressively disable the ones you find non necessary for your project
on an existing project using Pylint, when you want to test what extra checks
can be performed by this tool, and see if it can spot bugs you missed so far

The following .pylintrc configuration file is my attempt to setup a "strict" mode:

[MESSAGES CONTROL]
# Enable some checkers that are not activated by default:
enable = bad-inline-option, deprecated-pragma, file-ignored, use-symbolic-message-instead, useless-suppression

# Include some helpful details on errors messages for naming rules:
include-naming-hint = yes

[MASTER]
# Informational messages ("I") should make Pylint execution fail (non-0 program return code):
fail-on = I
# Enable many optional extensions:
load-plugins = pylint.extensions.bad_builtin,
               pylint.extensions.code_style,
               pylint.extensions.comparison_placement,
               pylint.extensions.consider_refactoring_into_while_condition,
               pylint.extensions.docparams,
               pylint.extensions.dunder,
               pylint.extensions.eq_without_hash,
               pylint.extensions.for_any_all,
               pylint.extensions.magic_value,
               pylint.extensions.mccabe,
               pylint.extensions.no_self_use,
               pylint.extensions.overlapping_exceptions,
               pylint.extensions.private_import,
               pylint.extensions.redefined_loop_name,
               pylint.extensions.redefined_variable_type,
               pylint.extensions.set_membership,
               pylint.extensions.typing,
               pylint.extensions.while_used,

[STRING_CONSTANT]
# Doc: https://pylint.pycqa.org/en/latest/user_guide/messages/warning/implicit-str-concat.html
check-quote-consistency = yes

[VARIABLES]
allow-global-unused-variables = no

I recently used this approach on fpdf2: PR #780 Hardening Pylint config

Related configuration files in fpdf2:

I hope this .pylintrc could be useful to others!

You can drop a comment if this article was helpful to you 😊

fpdf2 latest new features

Lucas Cimon — Mon, 03 Apr 2023 14:43:13 +0000

fpdf2 is a simple & fast PDF creation library for Python:

https://github.com/pyfpdf/fpdf2/ - Doc: https://pyfpdf.github.io/fpdf2/

As we just released a new version, v2.7,
this is the time to mention some recent additions! 😊

This article will present some of the major features introduced between v2.5.3 & v2.7.3 of fpdf2:

Tables
Images & shapes
Fonts & text
Signing & encrypting documents
Annotations
Documentation & translated tutorials
What's coming next?

Read the full article there: https://chezsoi.org/lucas/blog/fpdf2-latest-new-features.html

fpdf2.5.2 : SVG support and comparison with borb

Lucas Cimon — Sun, 24 Apr 2022 11:12:01 +0000

fpdf2 is a simple & fast PDF creation library for Python that I have been maintaining since mid-2020.

https://github.com/pyfpdf/fpdf2/ - Doc: https://pyfpdf.github.io/fpdf2/

In this article, I'm going to present some of the new features that landed since my last post on the subject, including SVG support.

I will also perform a quick comparison with the borb library.

More after the jump: https://chezsoi.org/lucas/blog/fpdf2-5-2-svg-support-and-borb.html

Bonnes pratiques Gitlab CI

Lucas Cimon — Thu, 29 Jul 2021 12:18:13 +0000

À E-voyageurs Technologies, je travaille au sein d'une équipe en charge de l'usine logicielle, qui administre depuis plusieurs années une instance Gitlab self-hosted.

Cet article contient quelques-unes de nos recommandations à l'intention des utilisateurs de notre Gitlab, ayant pour but à la fois améliorer les performances de leurs pipelines, et limiter leur impact en termes de ressources sur cette instance Gitlab partagée entre des dizaines d'équipes. Un dernier volet rassemble quelques points de sécurité.

https://chezsoi.org/lucas/blog/bonnes-pratiques-gitlab-ci.html

New release for fpdf2

Lucas Cimon — Sun, 13 Jun 2021 13:50:32 +0000

This release brings basic Markdown styling, JPEG storage improvements, presentation transitions and many other features!

More details, including how to convert GIFs to PDFs, in this announcement blog post: https://chezsoi.org/lucas/blog/fpdf2-4-0-and-converting-gifs-to-pdfs.html