I have organized the generation of my markdown documents that way:

我以这种方式组织了我的降价文档的生成：

export APPRENTISSAGE_HOME=the folder where my documents are

/apprentissage
    gen.sh : a script running a command: 'pandoc -f markdown-implicit_figures 
             --syntax-definition=$APPRENTISSAGE_HOME/java.xml 
             "${source}".md -o "${source}".pdf'
        → gen e020-analyse generates a e020-analyse.pdf from a e020-analyse.md file

    apprentissage-header-include.tex: \usepackage, \DeclareUnicodeCharacter directives
        \usepackage{fvextra}
        \usepackage{mathtools}
        \DeclareUnicodeCharacter{3C3}{\text{$\sigma$}}
        \DisableLigatures[-]{}

    apprentissage-include.tex: definitions of custom commands or colors
        \DefineVerbatimEnvironment{Highlighting}{Verbatim}{breaklines,breakanywhere,breaksymbol=,breakanywheresymbolpre=,commandchars=\\\{\}}
        \setlength{\abovedisplayskip}{3pt}

        \newcommand{\suspect}[1]{\textbf{\textcolor{gray}{#1}}}
        \newcommand{\fonction}[1]{\begingroup\small\mauve{\texttt{#1}}\endgroup}
        \renewcommand{\abstractname}{Contenu}

        \definecolor{cadreCodeBackground}{RGB}{130,124,106}
        \newcommand{\cadreCode}{\renewtcolorbox{quote}{colback=cadreCodeBackground!5!white,arc=0pt,outer arc=0pt,boxrule=0pt,lowerbox=invisible,code={\tcbset{enlarge left by=0cm}}}}

    /informatique this folder and the next have about 30 markdowns file inside
    /mathematiques
    /statistiques
    /analyse_spatiale
    /territoire

To work, each of my markdown files are starting the way below, and I would like to improve the the part between the --- and ---:

为了工作，我的每个降价文件都是从下面的方式开始的，我想改进---和---之间的部分：

---
header-includes:
  \input{$APPRENTISSAGE_HOME/apprentissage-header-include}
  \input{$APPRENTISSAGE_HOME/apprentissage-include}

title: "e020. Analyse textuelle"
subtitle: ""
category: analyse textuelle

keywords:
- TF
- term frequency
- IDF
- inverse difference frequency
- mot
- ponctuation
- token

author: Marc Le Bihan
geometry: margin=1.5cm
fontsize: 12pt
output: pdf
classoption: fleqn
urlcolor: blue
---

Indexation des mots. La qualité de la recherche vient par l'emploi de deux indicateurs :

   - __TF__ (Term Frequency) : fréquence des mots.
[...]

In the part between --- separators, I have two types of contents:

在分隔符之间的部分，我有两种类型的内容：

• One that is "static" to me: author, geometry, fontsize, output, classoption, urlcolor and

  对我来说是“静态”的：author、geometry、fontsize、output、classoption、urlcolor和

  
  

  header-includes:
     \input{$APPRENTISSAGE_HOME/apprentissage-header-include}
     \input{$APPRENTISSAGE_HOME/apprentissage-include}
  

  
  

  never changes, at the top of files.

  永远不会更改，位于文件顶部。

  
  



• title, subtitle, category, keywords that are changing from file to file.

  标题、副标题、类别、关键字随文件的变化而变化。

  
  

I would like to factorize more that part. Because it happens (whatever I've said) that I have sometimes to change that "static" part... And then, it's long to achieve: all the markdown files need edition and to have their pdf regenerated.

我想把那个部分分解得更多。因为有时候（不管我说了什么）我不得不改变“静态”的部分。。。然后，要实现这一点还需要很长时间：所有的降价文件都需要编辑并重新生成pdf。

• May I dispatch some of the keywords author, geometry, fontsize, output, classoption, urlcolor or header-includes: in one of the include .tex file I've created, and how?

  我可以在我创建的include.tex文件中发送一些关键字author、geometry、fontsize、output、classoption、urlcolor或header-includs吗？如何发送？

  
  



• Or should I improve my gen.sh running pandoc some way?

  或者我应该以某种方式改进我的潘多克一代吗？

  
  



• Can another feature of Latex I haven't used yet, help me?

  乳胶的另一个功能我还没有用过，能帮我吗？

  
  

A part of markdown document options can be transferred to a metadata.json file that Pandoc can include.

降价文档选项的一部分可以传输到Pandoc可以包含的metadata.json文件中。

geometry: margin=1.5cm
fontsize: 12pt
classoption: fleqn
urlcolor: blue

metadata.json:

metadata.json：

{
  "geometry": "margin=4cm",
  "fontsize": "12pt",
  "classoption": "fleqn",
  "urlcolor": "blue"
}

with the Pandoc command adding a --metadata-file inside:

Pandoc命令在其中添加一个--metadata文件：

pandoc -f markdown-implicit_figures \
  --syntax-definition=$APPRENTISSAGE_HOME/java.xml \
  --metadata-file=$APPRENTISSAGE_HOME/metadata.json \
  s600-la_fouille_de_données.md -o s600-la_fouille_de_données.pdf

I can also move my header includes:

我还可以移动我的页眉，包括：

header-includes:
  \input{$APPRENTISSAGE_HOME/apprentissage-header-include}
  \input{$APPRENTISSAGE_HOME/apprentissage-include}

to the Pandoc command:

到Pandoc命令：

pandoc -f markdown-implicit_figures \
  --syntax-definition=$APPRENTISSAGE_HOME/java.xml \
  --metadata-file=$APPRENTISSAGE_HOME/metadata.json \
  --include-in-header=$APPRENTISSAGE_HOME/apprentissage-header-include.tex \
  --include-in-header=$APPRENTISSAGE_HOME/apprentissage-include.tex \
  s600-la_fouille_de_données.md -o s600-la_fouille_de_données.pdf

This leaves me with a markdown preamble completely specific to its purpose, I was searching for.

这给我留下了一个完全针对我所寻找的目的的降价序言。

---
title: "e020. Analyse textuelle"
subtitle: ""
category: analyse textuelle

keywords:
- TF
- term frequency
- IDF
- inverse difference frequency
- mot
- ponctuation
- token

author: Marc Le Bihan
---