[Hide]

sgml
#define sgml: \
I---------------------------------------------------\
I---------------------------------------------------\
I---------------------------------------------------\
I      /$$$$$$   /$$$$$$  /$$      /$$ /$$          \
I     /$$__  $$ /$$__  $$| $$$    /$$$| $$          \
I    | $$  \__/| $$  \__/| $$$$  /$$$$| $$          \
I    |  $$$$$$ | $$ /$$$$| $$ $$/$$ $$| $$          \
I     \____  $$| $$|_  $$| $$  $$$| $$| $$          \
I     /$$  \ $$| $$  \ $$| $$\  $ | $$| $$          \
I    |  $$$$$$/|  $$$$$$/| $$ \/  | $$| $$$$$$$$    \
I     \______/  \______/ |__/     |__/|________/    \
I---------------------------------------------------\
I---------------------------------------------------\
I---------------------------------------------------I

    • "Standard Generalized Markup Language"
    • the forefather of many markup languages
    • used for describing structured documents
    • hierarchical by nature
    • has only historical significance, to explain the relationship
      between modern markup languages
    { // Evolution --->
                           ┌──▶ numerous GUI specification languages
                           │
                ┌──▶ XML ──┴──▶ mathML
                │      └───┐
        SGML ───┤          └─┐
                │            ├──▶ XHTML
                └──▶ HTML ───┘
    }

    Tags:
    Tags:
        <tag-name attributes*>
    (contents)
</tag-name>
<!-- or -->
<tag-name />

        • the smallest building block of a document
        • tags are nested inside one another to represent hierarchy
        • tags must be closed before their parents are closed;
          no intersection type structure is allowed
        • formally, an element is an instance of a tag
        • however tag is also the term used for describing the syntax element
          delimited by '<' and '>'
        • the tag that marks the beginning of an element is called an opening tag
        • the tag that marks the end of an element is called a closing tag
        • paired elements consist of an opening tag and a closing tag and
          optionally further markdown between the two
        • unpaired tags consist if a single self-closing tag;
          they can have no contents { forced line break }
        • the terms parent, chidren and sibling are used to express
          the relationship between elements
        <hello-world> My best greetings! </hello-world>
      A              A                  A
       \-------------|------------------|-------- An opening tag          \
                      \-----------------|-------- Text inside the element  } A paired element.
                                         \------- A closing tag           /
<example />
   A       
   |       
An unpaired element.

        <a> <!-- parent of elements 'b' and 'c' -->
    <b/> <!-- child of 'a', sibling of 'c' -->
    <c/> <!-- child of 'a', sibling of 'b' -->
</a>

        Attributes:
            • live inside an opening tag
            • optionally can have a value
            • contains metadata about the element
            • space separated from the tag type and other attributes
            <hello-world id="HW"> </hello-world>
                A
                |
   id attribute with the value HW,
  (some variation of this is usually
     supported to name elements)


    DTD:
    DTD:
        
            https://www.htmlhelp.com/design/dtd/customdtd.html
        • "Document Type Definition"
        • tag definition language
        • the tags of SGML are user defined, provided in the DTD format
        • standard HTML structures are specified in DTD

    XML:
    XML:
        • "Extensible Markup Language"
        • provides the freedom of defining your own tags and attributes
        • a stricter alternative to SGML which is easier to parse
        • Microsoft uses it extensively { WPF; MSBuild; }
        • since the exact syntax of any XML document requires extra context,
          there is not much to say about XML in a vacuum

    XPATH:
    XPATH:
        • "XML Path Language"
        • technology for locating information inside XML
        •  suitable for SGML; HTML supports it too
        • by design, similar to UNIX file paths,
          where tags behave like folders;
          this is a perfectly clear abstraction since both describe trees
        
            • sometimes useful for web-scrapping
        — examples:
            • assume the following document
            <data>
    <employees>
        <employee id=12>
            <first-name>John</first-name>
            <last-name>Doe</last-name>
        </employee>
        <employee id=4>
            <first-name>Anon</first-name>
            <last-name>Anonson</last-name>
        </employee>
    </employees>
</data>

            • the tag names behave like folder names
            • nodes are separated by '/'s
            /data/employees

            • if there are multiple matches inside the specified tag,
              they behave like an array with 1-based indexing
            # selects <employee id=12>
#  NOTE: employee[0] does not exist
/data/employees/employee[1]

            • one can also filter by tag value
            # selects <employee id=4>
/data/employees/employee[first-name = "Anon"]

            • or attribute
             #selects <employee id=4>
/data/employees/employee[@id = "4"]

            • or access the parent
            #selects <employees>
/data/employees/employee[first-name = "Anon"]/..