feat: Start work on !ELEMENT

gleam.toml

@@ -4,9 +4,9 @@ version = "1.0.0"
 # Fill out these fields if you intend to generate HTML documentation or publish
 # your project to the Hex package manager.
 #
-# description = ""
-# licences = ["Apache-2.0"]
-# repository = { type = "github", user = "", repo = "" }
+description = "Gleam XML parser"
+licences = ["Apache-2.0"]
+repository = { type = "gitea", host = "git.pendleton.ie", user = "pendletong", repo = "glxml" }
 # links = [{ title = "Website", href = "" }]
 #
 # For a full reference of all the available options, you can have a look at

src/glxml.gleam

@@ -6,8 +6,33 @@ import gleam/option.{type Option, None, Some}
 import gleam/result
 import gleam/string
 
+pub type Content {
+  Empty
+  Any
+  Mixed(content: List(String))
+  Choice(content: ContentParticle)
+}
+
 pub type Declaration {
-  Declaration(versioninfo: String, encoding: String, standalone: Bool)
+  XMLDecl(versioninfo: String, encoding: String, standalone: Bool)
+  GEntityDecl
+  PEntityDecl(name: String, decl: String)
+  ElementDecl(name: String, content: Content)
+  AttListDecl
+  NotationDecl
+}
+
+pub type Optional {
+  One
+  OneOrMore
+  ZeroOrMore
+  ZeroOrOne
+}
+
+pub type ContentParticle {
+  ElParticle(name: String, optional: Optional)
+  ChoiceParticle(choices: List(ContentParticle), optional: Optional)
+  SeqParticle(seq: List(ContentParticle), optional: Optional)
 }
 
 pub type Entity {
@@ -16,8 +41,17 @@ pub type Entity {
   PublicExternalEntity(literal: String, pubidliteral: String)
 }
 
+pub type ExternalID {
+  SystemID(system_literal: String)
+  PublicID(system_literal: String, public_literal: String)
+}
+
 pub type DocType {
-  DocType(name: String, entities: dict.Dict(String, Entity))
+  DocType(
+    name: String,
+    external_id: Option(ExternalID),
+    entities: dict.Dict(String, Entity),
+  )
 }
 
 pub type Document {
@@ -35,16 +69,29 @@ pub type Attribute {
 pub type Element {
   EmptyElem(name: String, attrs: List(Attribute))
   Element(name: String, attrs: List(Attribute), elements: List(Element))
+  Text(content: String)
+  Comment(content: String)
+  CData(content: String)
+  PI(name: String, content: String)
+  Whitespace
 }
 
 pub fn main() {
   parse_document(
-    "<?xml version=\"1.1\" encoding='UTF-8'?>\r\n   <!-- hello-world -->   \n<a attr='ha&#x20;&#38;#38;ha' battr='baba' ref='&amp;'/>",
+    "<?xml version=\"1.1\"?>\r\n<!DOCTYPE el [\r\n<!ELEMENT div1 (head, (p | list | note)*, div2*)>\r\n]>",
+    //"<?xml version=\"1.1\" encoding='UTF-8'?>\r\n   <!-- hello-world -->   \n<b blah:test='1'><a attr='ha&#x20;&#38;#38;ha' battr='baba' ref='&amp;'/><!-- ma comment --><![CDATA[Testing&&<haha>]]><?test asuhashd ?></b>",
+  // "<doc>
+  //     "<doc>
+  // <A a=\"asdf>'&#34;>
+  // asdf
+  // 	?>%\"/>
+  // <A a='\"\">&#39;&#34;'/>
+  // </doc>",
   )
   |> echo
 }
 
-pub fn default_entities() -> dict.Dict(String, Entity) {
+fn default_entities() -> dict.Dict(String, Entity) {
   dict.from_list([
     #("lt", InternalEntity("&#60;")),
     #("gt", InternalEntity("&#62;")),
@@ -70,7 +117,10 @@ fn parse_element(
   doc: String,
   doctype: Option(DocType),
 ) -> Result(#(Element, String), Nil) {
-  try_parsers([parse_empty_elem(_, doctype)], doc)
+  try_parsers(
+    [parse_empty_elem(_, doctype), parse_tagged_elem(_, doctype)],
+    doc,
+  )
 }
 
 fn parse_empty_elem(
@@ -91,6 +141,155 @@ fn parse_empty_elem(
   }
 }
 
+fn parse_tagged_elem(
+  doc: String,
+  doctype: Option(DocType),
+) -> Result(#(Element, String), Nil) {
+  case doc {
+    "<" <> tail -> {
+      use #(name, doc) <- result.try(parse_name(tail))
+      use #(attrs, doc) <- result.try(parse_attributes(doc, doctype, []))
+      let doc = trim_space(doc)
+      case doc {
+        ">" <> tail -> {
+          use #(content, doc) <- result.try(parse_content(tail, doctype, []))
+
+          case doc {
+            "</" <> tail -> {
+              use #(close_name, doc) <- result.try(parse_name(tail))
+              let doc = trim_space(doc)
+              case doc {
+                ">" <> tail -> {
+                  use <- bool.guard(
+                    when: name != close_name,
+                    return: Error(Nil),
+                  )
+
+                  Ok(#(Element(name, attrs, content), tail))
+                }
+                _ -> Error(Nil)
+              }
+            }
+            _ -> Error(Nil)
+          }
+        }
+        _ -> Error(Nil)
+      }
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn parse_content(
+  doc: String,
+  doctype: Option(DocType),
+  content: List(Element),
+) -> Result(#(List(Element), String), Nil) {
+  use #(chardata, doc) <- result.try(parse_chardata(doc, doctype, ""))
+  let content = case chardata {
+    "" -> content
+    _ -> [Text(chardata), ..content]
+  }
+
+  case doc {
+    "<" <> _ -> {
+      case
+        try_parsers(
+          [
+            parse_element(_, doctype),
+            parse_comment,
+            parse_cdata,
+            parse_pi,
+          ],
+          doc,
+        )
+      {
+        Ok(#(element, doc)) -> parse_content(doc, doctype, [element, ..content])
+        Error(_) -> Ok(#(list.reverse(content), doc))
+      }
+    }
+    _ -> Ok(#(list.reverse(content), doc))
+  }
+}
+
+fn parse_pi(doc: String) -> Result(#(Element, String), Nil) {
+  case doc {
+    "<?" <> tail -> {
+      use #(name, doc) <- result.try(parse_name(tail))
+      use <- bool.guard(
+        when: string.lowercase(name) == "xml",
+        return: Error(Nil),
+      )
+      let doc = trim_space(doc)
+      use #(content, doc) <- result.try(parse_pi_content(doc, ""))
+      case doc {
+        "?>" <> tail -> Ok(#(PI(name, content), tail))
+        _ -> Error(Nil)
+      }
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn parse_pi_content(doc: String, pi: String) -> Result(#(String, String), Nil) {
+  case doc {
+    "?>" <> _ -> Ok(#(pi, doc))
+    "" -> Error(Nil)
+    _ -> {
+      case parse_char(doc) {
+        Ok(#(char, doc)) -> parse_pi_content(doc, pi <> char)
+        Error(_) -> Ok(#(pi, doc))
+      }
+    }
+  }
+}
+
+fn parse_cdata(doc: String) -> Result(#(Element, String), Nil) {
+  case doc {
+    "<![CDATA[" <> tail -> {
+      use #(cdata, doc) <- result.try(do_parse_cdata(tail, ""))
+
+      Ok(#(CData(cdata), doc))
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn do_parse_cdata(doc: String, cdata: String) -> Result(#(String, String), Nil) {
+  case doc {
+    "]]>" <> tail -> Ok(#(cdata, tail))
+    "" -> Error(Nil)
+    _ -> {
+      case parse_char(doc) {
+        Ok(#(char, doc)) -> do_parse_cdata(doc, cdata <> char)
+        _ -> Error(Nil)
+      }
+    }
+  }
+}
+
+fn parse_chardata(
+  doc: String,
+  doctype: Option(DocType),
+  chardata: String,
+) -> Result(#(String, String), Nil) {
+  case doc {
+    "]]>" <> _ -> Error(Nil)
+    "<" <> _ -> Ok(#(chardata, doc))
+    "&" <> _ -> {
+      use #(refval, doc) <- result.try(parse_reference(doc, doctype))
+
+      parse_chardata(doc, doctype, chardata <> refval)
+    }
+
+    "" -> Ok(#("", ""))
+    _ -> {
+      let assert Ok(#(char, tail)) = string.pop_grapheme(doc)
+      parse_chardata(tail, doctype, chardata <> char)
+    }
+  }
+}
+
 fn parse_attributes(
   doc: String,
   doctype: Option(DocType),
@@ -106,7 +305,7 @@ fn parse_attribute(
   doc: String,
   doctype: Option(DocType),
 ) -> Result(#(Attribute, String), Nil) {
-  let doc = trim_space(doc)
+  use doc <- result.try(trim_mandatory_space(doc))
   use #(name, doc) <- result.try(parse_name(doc))
   case doc {
     "=" <> tail -> {
@@ -203,7 +402,7 @@ fn parse_reference(
         }
       }
     }
-    "&" as char <> tail -> {
+    "&" <> tail -> {
       use #(name, doc) <- result.try(parse_name(tail))
 
       case doc {
@@ -223,7 +422,7 @@ fn process_reference(
   doctype: Option(DocType),
 ) -> Result(String, Nil) {
   case doctype {
-    Some(DocType(_, entities)) -> {
+    Some(DocType(_, _, entities)) -> {
       get_reference(entities, ref)
     }
     None -> {
@@ -267,41 +466,512 @@ fn do_parse_name(doc: String, name: String) -> Result(#(String, String), Nil) {
 fn parse_prolog(
   doc: String,
 ) -> Result(#(Declaration, Option(DocType), String), Nil) {
-  let #(decl, doc) = case parse_decl(doc) {
-    Ok(#(decl, doc)) -> #(decl, doc)
-    _ -> #(Declaration("1.0", "UTF-8", False), doc)
-  }
+  let #(decl, doc) =
+    parse_decl(doc) |> result.unwrap(#(XMLDecl("1.0", "UTF-8", False), doc))
+
   let doc = parse_misc(doc)
 
-  Ok(#(decl, None, doc))
+  let #(doctype, doc) =
+    parse_doctype(doc)
+    |> result.map(fn(d) { #(Some(d.0), d.1) })
+    |> result.unwrap(#(None, doc))
+
+  Ok(#(decl, doctype, doc))
+}
+
+fn parse_doctype(doc: String) -> Result(#(DocType, String), Nil) {
+  case doc {
+    "<!DOCTYPE" <> tail -> {
+      use doc <- result.try(trim_mandatory_space(tail))
+      use #(name, doc) <- result.try(parse_name(doc))
+
+      let #(external_id, doc) =
+        parse_external_id(doc) |> result.unwrap(#(None, doc))
+
+      let doc = trim_space(doc)
+
+      let #(int_subset, doc) =
+        parse_int_subset(doc) |> result.unwrap(#([], doc))
+
+      case doc {
+        ">" <> tail -> Ok(#(DocType(name, external_id, dict.new()), tail))
+        _ -> Error(Nil)
+      }
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn parse_int_subset(doc: String) -> Result(#(List(Declaration), String), Nil) {
+  let doc = trim_space(doc)
+  case doc {
+    "[" <> tail -> {
+      use #(decl_list, doc) <- result.try(do_parse_int_subset(tail, []))
+      case doc {
+        "]" <> tail -> {
+          Ok(#(decl_list, tail))
+        }
+        _ -> Error(Nil)
+      }
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn do_parse_int_subset(
+  doc: String,
+  decl_list: List(Declaration),
+) -> Result(#(List(Declaration), String), Nil) {
+  let doc = trim_space(doc)
+  case doc {
+    "%" <> tail -> {
+      use #(name, doc) <- result.try(parse_name(tail))
+      case doc {
+        ";" <> tail -> {
+          case get_entity_replacement(name, decl_list) {
+            Some(decl) -> {
+              do_parse_int_subset(decl <> tail, decl_list)
+            }
+            _ -> Error(Nil)
+          }
+        }
+        _ -> Error(Nil)
+      }
+    }
+    _ -> {
+      case
+        try_parsers(
+          [
+            parse_elementdecl,
+          ],
+          doc,
+        )
+      {
+        Ok(#(decl, doc)) -> {
+          do_parse_int_subset(doc, [decl, ..decl_list])
+        }
+        Error(_) -> Ok(#(list.reverse(decl_list), doc))
+      }
+    }
+  }
+}
+
+fn parse_elementdecl(doc: String) -> Result(#(Declaration, String), Nil) {
+  case doc {
+    "<!ELEMENT" <> tail -> {
+      use doc <- result.try(trim_mandatory_space(tail))
+      use #(name, doc) <- result.try(parse_name(doc))
+      use doc <- result.try(trim_mandatory_space(doc))
+      case doc {
+        "EMPTY" <> tail -> {
+          let doc = trim_space(tail)
+          case doc {
+            ")" <> tail -> {
+              Ok(#(ElementDecl(name, Empty), tail))
+            }
+            _ -> Error(Nil)
+          }
+        }
+        "ANY" <> tail -> {
+          let doc = trim_space(tail)
+          case doc {
+            ")" <> tail -> {
+              Ok(#(ElementDecl(name, Any), tail))
+            }
+            _ -> Error(Nil)
+          }
+        }
+        "(" <> _ -> {
+          try_parsers([parse_mixed(name, _), parse_children(name, _)], doc)
+        }
+        _ -> Error(Nil)
+      }
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn parse_children(
+  name: String,
+  doc: String,
+) -> Result(#(Declaration, String), Nil) {
+  use #(children, doc) <- result.try(try_parsers([parse_choice, parse_seq], doc))
+
+  let #(children, doc) = case doc {
+    "?" <> tail -> {
+      #(set_optional(children, ZeroOrOne), tail)
+    }
+    "*" <> tail -> {
+      #(set_optional(children, ZeroOrMore), tail)
+    }
+    "+" <> tail -> {
+      #(set_optional(children, OneOrMore), tail)
+    }
+    _ -> {
+      #(children, doc)
+    }
+  }
+  children |> echo
+  doc |> echo
+  todo
+}
+
+fn parse_choice(doc: String) -> Result(#(ContentParticle, String), Nil) {
+  case doc {
+    "(" <> tail -> {
+      let doc = trim_space(tail)
+      use #(cp, doc) <- result.try(parse_cp(doc))
+      use #(cps, doc) <- result.try(case do_parse_choice(doc, [cp]) {
+        Ok(#(ChoiceParticle([], _), _))
+        | Ok(#(ChoiceParticle([_], _), _))
+        | Error(_) -> Error(Nil)
+        Ok(#(cps, doc)) -> Ok(#(cps, doc))
+      })
+      let doc = trim_space(doc)
+      case doc {
+        ")" <> tail -> Ok(#(cps, tail))
+        _ -> Error(Nil)
+      }
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn do_parse_choice(
+  doc: String,
+  acc: List(ContentParticle),
+) -> Result(#(ContentParticle, String), Nil) {
+  let doc = trim_space(doc)
+  case doc {
+    "|" <> tail -> {
+      let doc = trim_space(tail)
+      use #(cp, doc) <- result.try(parse_cp(doc))
+      do_parse_choice(doc, [cp, ..acc])
+    }
+    _ -> Ok(#(ChoiceParticle(list.reverse(acc), One), doc))
+  }
+}
+
+fn parse_seq(doc: String) -> Result(#(ContentParticle, String), Nil) {
+  case doc {
+    "(" <> tail -> {
+      let doc = trim_space(tail)
+      use #(cp, doc) <- result.try(parse_cp(doc))
+      let #(cps, doc) = case do_parse_seq(doc, [cp]) {
+        Ok(#(cps, doc)) -> #(cps, doc)
+        Error(_) -> #(SeqParticle([cp], One), doc)
+      }
+      let doc = trim_space(doc)
+      case doc {
+        ")" <> tail -> Ok(#(cps, tail))
+        _ -> Error(Nil)
+      }
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn do_parse_seq(
+  doc: String,
+  acc: List(ContentParticle),
+) -> Result(#(ContentParticle, String), Nil) {
+  let doc = trim_space(doc)
+  case doc {
+    "," <> tail -> {
+      let doc = trim_space(tail)
+      use #(cp, doc) <- result.try(parse_cp(doc))
+      do_parse_seq(doc, [cp, ..acc])
+    }
+    _ -> Ok(#(SeqParticle(list.reverse(acc), One), doc))
+  }
+}
+
+fn parse_cp(doc: String) -> Result(#(ContentParticle, String), Nil) {
+  use #(el, doc) <- result.try(try_parsers(
+    [
+      fn(doc) {
+        use #(name, doc) <- result.try(parse_name(doc))
+        Ok(#(ElParticle(name, One), doc))
+      },
+      parse_choice,
+      parse_seq,
+    ],
+    doc,
+  ))
+
+  case doc {
+    "?" <> tail -> {
+      Ok(#(set_optional(el, ZeroOrOne), tail))
+    }
+    "*" <> tail -> {
+      Ok(#(set_optional(el, ZeroOrMore), tail))
+    }
+    "+" <> tail -> {
+      Ok(#(set_optional(el, OneOrMore), tail))
+    }
+    _ -> {
+      Ok(#(el, doc))
+    }
+  }
+}
+
+fn set_optional(el: ContentParticle, optional: Optional) -> ContentParticle {
+  case el {
+    ChoiceParticle(choices, _) -> ChoiceParticle(choices:, optional:)
+    ElParticle(name, _) -> ElParticle(name:, optional:)
+    SeqParticle(seq, _) -> SeqParticle(seq:, optional:)
+  }
+}
+
+fn parse_mixed(name: String, doc: String) -> Result(#(Declaration, String), Nil) {
+  case doc {
+    "(" <> tail -> {
+      let doc = trim_space(tail)
+      case doc {
+        "#PCDATA" <> tail -> {
+          use #(els, doc) <- result.try(parse_mixed_elements(tail))
+          let doc = trim_space(doc)
+          case els {
+            [] -> {
+              case doc {
+                ")*" <> tail ->
+                  Ok(#(ElementDecl(name, Mixed(["#PCDATA"])), tail))
+                _ -> Error(Nil)
+              }
+            }
+            _ -> {
+              case doc {
+                ")" <> tail ->
+                  Ok(#(ElementDecl(name, Mixed(["#PCDATA", ..els])), tail))
+                _ -> Error(Nil)
+              }
+            }
+          }
+        }
+        _ -> Error(Nil)
+      }
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn parse_mixed_elements(doc: String) -> Result(#(List(String), String), Nil) {
+  Ok(
+    parse_multiple_to_list(
+      doc,
+      fn(doc) {
+        let doc = trim_space(doc)
+        case doc {
+          "|" <> tail -> {
+            let doc = trim_space(tail)
+            parse_name(doc)
+          }
+          _ -> Error(Nil)
+        }
+      },
+      [],
+    ),
+  )
+}
+
+fn get_entity_replacement(
+  entity: String,
+  decl_list: List(Declaration),
+) -> Option(String) {
+  list.find_map(decl_list, fn(decl) {
+    case decl {
+      PEntityDecl(name, decl) if name == entity -> Ok(Some(decl))
+      _ -> Error(Nil)
+    }
+  })
+  |> result.unwrap(None)
+}
+
+fn parse_external_id(doc: String) -> Result(#(Option(ExternalID), String), Nil) {
+  use doc <- result.try(trim_mandatory_space(doc))
+  case doc {
+    "SYSTEM" <> tail -> {
+      use doc <- result.try(trim_mandatory_space(tail))
+      use #(system_literal, doc) <- result.try(parse_system_literal(
+        doc,
+        None,
+        "",
+      ))
+      Ok(#(Some(SystemID(system_literal:)), doc))
+    }
+    "PUBLIC" <> tail -> {
+      use doc <- result.try(trim_mandatory_space(tail))
+      use #(public_literal, doc) <- result.try(parse_public_literal(
+        doc,
+        None,
+        "",
+      ))
+      use doc <- result.try(trim_mandatory_space(doc))
+      use #(system_literal, doc) <- result.try(parse_system_literal(
+        doc,
+        None,
+        "",
+      ))
+      Ok(#(Some(PublicID(system_literal:, public_literal:)), doc))
+    }
+    _ -> Error(Nil)
+  }
+}
+
+fn parse_public_literal(
+  doc: String,
+  quote: Option(String),
+  literal: String,
+) -> Result(#(String, String), Nil) {
+  case doc, quote {
+    "\"" as q <> tail, None | "'" as q <> tail, None ->
+      parse_public_literal(tail, Some(q), "")
+    "", _ -> Error(Nil)
+    _, None -> Error(Nil)
+    "\"" <> tail, Some("\"") -> Ok(#(literal, tail))
+    "'" <> tail, Some("'") -> Ok(#(literal, tail))
+    " " as char <> tail, Some(_)
+    | "\r" as char <> tail, Some(_)
+    | "\n" as char <> tail, Some(_)
+    | "0" as char <> tail, Some(_)
+    | "1" as char <> tail, Some(_)
+    | "2" as char <> tail, Some(_)
+    | "3" as char <> tail, Some(_)
+    | "4" as char <> tail, Some(_)
+    | "5" as char <> tail, Some(_)
+    | "6" as char <> tail, Some(_)
+    | "7" as char <> tail, Some(_)
+    | "8" as char <> tail, Some(_)
+    | "9" as char <> tail, Some(_)
+    | "a" as char <> tail, Some(_)
+    | "b" as char <> tail, Some(_)
+    | "c" as char <> tail, Some(_)
+    | "d" as char <> tail, Some(_)
+    | "e" as char <> tail, Some(_)
+    | "f" as char <> tail, Some(_)
+    | "g" as char <> tail, Some(_)
+    | "h" as char <> tail, Some(_)
+    | "i" as char <> tail, Some(_)
+    | "j" as char <> tail, Some(_)
+    | "k" as char <> tail, Some(_)
+    | "l" as char <> tail, Some(_)
+    | "m" as char <> tail, Some(_)
+    | "n" as char <> tail, Some(_)
+    | "o" as char <> tail, Some(_)
+    | "p" as char <> tail, Some(_)
+    | "q" as char <> tail, Some(_)
+    | "r" as char <> tail, Some(_)
+    | "s" as char <> tail, Some(_)
+    | "t" as char <> tail, Some(_)
+    | "u" as char <> tail, Some(_)
+    | "v" as char <> tail, Some(_)
+    | "w" as char <> tail, Some(_)
+    | "x" as char <> tail, Some(_)
+    | "y" as char <> tail, Some(_)
+    | "z" as char <> tail, Some(_)
+    | "A" as char <> tail, Some(_)
+    | "B" as char <> tail, Some(_)
+    | "C" as char <> tail, Some(_)
+    | "D" as char <> tail, Some(_)
+    | "E" as char <> tail, Some(_)
+    | "F" as char <> tail, Some(_)
+    | "G" as char <> tail, Some(_)
+    | "H" as char <> tail, Some(_)
+    | "I" as char <> tail, Some(_)
+    | "J" as char <> tail, Some(_)
+    | "K" as char <> tail, Some(_)
+    | "L" as char <> tail, Some(_)
+    | "M" as char <> tail, Some(_)
+    | "N" as char <> tail, Some(_)
+    | "O" as char <> tail, Some(_)
+    | "P" as char <> tail, Some(_)
+    | "Q" as char <> tail, Some(_)
+    | "R" as char <> tail, Some(_)
+    | "S" as char <> tail, Some(_)
+    | "T" as char <> tail, Some(_)
+    | "U" as char <> tail, Some(_)
+    | "V" as char <> tail, Some(_)
+    | "W" as char <> tail, Some(_)
+    | "X" as char <> tail, Some(_)
+    | "Y" as char <> tail, Some(_)
+    | "Z" as char <> tail, Some(_)
+    | "-" as char <> tail, Some(_)
+    | "(" as char <> tail, Some(_)
+    | ")" as char <> tail, Some(_)
+    | "+" as char <> tail, Some(_)
+    | "," as char <> tail, Some(_)
+    | "." as char <> tail, Some(_)
+    | "/" as char <> tail, Some(_)
+    | ":" as char <> tail, Some(_)
+    | "=" as char <> tail, Some(_)
+    | "?" as char <> tail, Some(_)
+    | ";" as char <> tail, Some(_)
+    | "!" as char <> tail, Some(_)
+    | "*" as char <> tail, Some(_)
+    | "#" as char <> tail, Some(_)
+    | "@" as char <> tail, Some(_)
+    | "$" as char <> tail, Some(_)
+    | "_" as char <> tail, Some(_)
+    | "%" as char <> tail, Some(_)
+    | "'" as char <> tail, Some("\"")
+    -> {
+      parse_public_literal(tail, quote, literal <> char)
+    }
+    _, _ -> Error(Nil)
+  }
+}
+
+fn parse_system_literal(
+  doc: String,
+  quote: Option(String),
+  literal: String,
+) -> Result(#(String, String), Nil) {
+  case doc, quote {
+    "\"" as q <> tail, None | "'" as q <> tail, None ->
+      parse_system_literal(tail, Some(q), "")
+    "", _ -> Error(Nil)
+    _, None -> Error(Nil)
+    "\"" <> tail, Some("\"") -> Ok(#(literal, tail))
+    "'" <> tail, Some("'") -> Ok(#(literal, tail))
+    _, _ -> {
+      let assert Ok(#(char, tail)) = string.pop_grapheme(doc)
+
+      parse_system_literal(tail, quote, literal <> char)
+    }
+  }
 }
 
 fn parse_misc(doc: String) -> String {
-  let #(_, doc) =
-    parse_multiple_optional(
+  case
+    try_parsers(
+      [
+        parse_comment,
+        parse_pi,
+        fn(doc) {
+          parse_space(doc)
+          |> result.map(fn(sp) { #(Whitespace, sp.1) })
+        },
+      ],
       doc,
-      try_parsers([parse_comment, parse_space], _),
-      "",
     )
-  doc
+  {
+    Ok(#(_element, doc)) -> parse_misc(doc)
+    Error(Nil) -> doc
+  }
 }
 
 fn parse_decl(doc: String) -> Result(#(Declaration, String), Nil) {
   case doc {
     "<?xml" <> tail -> {
       use #(versioninfo, doc) <- result.try(parse_versioninfo(tail))
-      let #(encoding, doc) = case parse_encodingdecl(doc) {
-        Ok(e) -> e
-        Error(_) -> #("", doc)
-      }
-      let #(standalone, doc) = case parse_standalone(doc) {
-        Ok(e) -> e
-        Error(_) -> #(False, doc)
-      }
+      let #(encoding, doc) =
+        parse_encodingdecl(doc) |> result.unwrap(#("", doc))
+      let #(standalone, doc) =
+        parse_standalone(doc) |> result.unwrap(#(False, doc))
 
       case trim_space(doc) {
         "?>" <> tail ->
-          Ok(#(Declaration(versioninfo:, encoding:, standalone:), tail))
+          Ok(#(XMLDecl(versioninfo:, encoding:, standalone:), tail))
         _ -> Error(Nil)
       }
     }
@@ -310,8 +980,8 @@ fn parse_decl(doc: String) -> Result(#(Declaration, String), Nil) {
 }
 
 fn parse_versioninfo(doc: String) -> Result(#(String, String), Nil) {
-  case trim_space(doc) {
-    "version=" <> tail -> {
+  case trim_mandatory_space(doc) {
+    Ok("version=" <> tail) -> {
       use #(version, doc) <- result.try(parse_version(tail))
       Ok(#(version, doc))
     }
@@ -351,8 +1021,8 @@ fn do_parse_version(
 }
 
 fn parse_encodingdecl(doc: String) -> Result(#(String, String), Nil) {
-  case trim_space(doc) {
-    "encoding=" <> tail -> {
+  case trim_mandatory_space(doc) {
+    Ok("encoding=" <> tail) -> {
       case tail {
         "\"" <> tail -> {
           use #(encoding, doc) <- result.try(parse_encoding(tail))
@@ -402,10 +1072,11 @@ fn parse_encoding(doc: String) -> Result(#(String, String), Nil) {
 }
 
 fn parse_standalone(doc: String) -> Result(#(Bool, String), Nil) {
-  case trim_space(doc) {
-    "standalone=\"yes\"" <> tail | "standalone='yes'" <> tail ->
+  case trim_mandatory_space(doc) {
+    Ok("standalone=\"yes\"" <> tail) | Ok("standalone='yes'" <> tail) ->
+      Ok(#(True, tail))
+    Ok("standalone=\"no\"" <> tail) | Ok("standalone='no'" <> tail) ->
       Ok(#(True, tail))
-    "standalone=\"no\"" <> tail | "standalone='no'" <> tail -> Ok(#(True, tail))
     _ -> Error(Nil)
   }
 }
@@ -426,7 +1097,7 @@ fn parse_digit(doc: String) -> Result(#(String, String), Nil) {
   }
 }
 
-pub fn parse_hex_digit(str: String) -> Result(#(String, String), Nil) {
+fn parse_hex_digit(str: String) -> Result(#(String, String), Nil) {
   case str {
     "0" as digit <> tail
     | "1" as digit <> tail
@@ -513,12 +1184,12 @@ fn parse_alpha(doc: String) -> Result(#(String, String), Nil) {
   }
 }
 
-fn parse_comment(doc: String) -> Result(#(String, String), Nil) {
+fn parse_comment(doc: String) -> Result(#(Element, String), Nil) {
   case doc {
     "<!--" <> tail -> {
       let #(comment, doc) = do_parse_comment(tail)
       case doc {
-        "-->" <> tail -> Ok(#(comment, tail))
+        "-->" <> tail -> Ok(#(Comment(comment), tail))
         _ -> Error(Nil)
       }
     }
@@ -634,6 +1305,18 @@ fn parse_name_char(doc: String) -> Result(#(String, String), Nil) {
   }
 }
 
+fn trim_mandatory_space(doc: String) -> Result(String, Nil) {
+  do_trim_mandatory_space(doc, True)
+}
+
+fn do_trim_mandatory_space(doc: String, first: Bool) -> Result(String, Nil) {
+  case parse_space(doc), first {
+    Ok(#(_, doc)), _ -> do_trim_mandatory_space(doc, False)
+    Error(_), True -> Error(Nil)
+    Error(_), False -> Ok(doc)
+  }
+}
+
 fn trim_space(doc: String) -> String {
   case parse_space(doc) {
     Ok(#(_, doc)) -> trim_space(doc)
@@ -665,7 +1348,7 @@ fn try_parsers(
   }
 }
 
-pub fn parse_multiple(
+fn parse_multiple(
   to_parse str: String,
   with to_run: fn(String) -> Result(#(String, String), Nil),
 ) -> Result(#(String, String), Nil) {
@@ -675,6 +1358,21 @@ pub fn parse_multiple(
   }
 }
 
+fn parse_multiple_to_list(
+  to_parse str: String,
+  with to_run: fn(String) -> Result(#(String, String), Nil),
+  acc ret: List(String),
+) -> #(List(String), String) {
+  case str {
+    "" -> #(list.reverse(ret), str)
+    _ ->
+      case to_run(str) {
+        Ok(#(r, rest)) -> parse_multiple_to_list(rest, to_run, [r, ..ret])
+        Error(_) -> #(list.reverse(ret), str)
+      }
+  }
+}
+
 fn parse_multiple_optional(
   to_parse str: String,
   with to_run: fn(String) -> Result(#(String, String), Nil),

test/oasis/e2.xml

@@ -0,0 +1,5 @@
+<!DOCTYPE el [
+<!ELEMENT el EMPTY>
+<!ATTLIST el at (one|two|two) #IMPLIED>
+]>
+<e1 at="two"/>

test/oasis/p01fail1.xml

@@ -0,0 +1,11 @@
+
+<?xml version="1.0"?>
+<doc>
+<a><b><c/></b></a>
+</doc>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>

test/oasis/p01fail2.xml

@@ -0,0 +1,10 @@
+<!--bad comment--><?xml version="1.0"?>
+<doc>
+<a><b><c/></b></a>
+</doc>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>

test/oasis/p01fail3.xml

@@ -0,0 +1,7 @@
+<doc/><bad/>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>

test/oasis/p01fail4.xml

@@ -0,0 +1 @@
+<doc>

test/oasis/p01pass1.xml

@@ -0,0 +1,3 @@
+<doc>
+<a><b><c/></b></a>
+</doc>

test/oasis/p01pass2.xml

@@ -0,0 +1,23 @@
+<?PI before document element?>
+<!-- comment after document element-->
+<?PI before document element?>
+<!-- comment after document element-->
+<?PI before document element?>
+<!-- comment after document element-->
+<?PI before document element?>
+<!DOCTYPE doc
+[
+<!ELEMENT doc ANY>
+<!ELEMENT a ANY>
+<!ELEMENT b ANY>
+<!ELEMENT c ANY>
+]>
+<doc>
+<a><b><c/></b></a>
+</doc>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>

test/oasis/p01pass3.xml

@@ -0,0 +1,9 @@
+<doc>
+<a><b><c/></b></a>
+</doc>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>
+<!-- comment after document element-->
+<?PI after document element?>

test/oasis/p03fail10.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail11.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail12.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail13.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail14.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail15.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail16.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail17.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail18.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail19.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail2.xml

@@ -0,0 +1 @@
+
<doc/>

test/oasis/p03fail20.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail21.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail22.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail23.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail24.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail25.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail26.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail27.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail28.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail29.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail3.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail4.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail5.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail7.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail8.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03fail9.xml

@@ -0,0 +1 @@
+<doc/>

test/oasis/p03pass1.xml

@@ -0,0 +1,2 @@
+	
+ <doc/>

test/oasis/p04fail1.xml

@@ -0,0 +1 @@
+<A@/>

test/oasis/p04fail2.xml

@@ -0,0 +1 @@
+<A#/>

test/oasis/p04fail3.xml

@@ -0,0 +1 @@
+<A$/>

test/oasis/p04pass1.xml

@@ -0,0 +1,6 @@
+<doc>
+<abcdefghijklmnopqrstuvwxyz/>
+<ABCDEFGHIJKLMNOPQRSTUVWXYZ/>
+<A01234567890/>
+<A.-:̀·/>
+</doc>

test/oasis/p05fail1.xml

@@ -0,0 +1 @@
+<0A/>

test/oasis/p05fail2.xml

@@ -0,0 +1 @@
+<.A/>

test/oasis/p05fail3.xml

@@ -0,0 +1 @@
+<-A/>

test/oasis/p05fail4.xml

@@ -0,0 +1 @@
+<̀A/>

test/oasis/p05fail5.xml

@@ -0,0 +1 @@
+<·A/>

test/oasis/p05pass1.xml

@@ -0,0 +1,8 @@
+<doc>
+<A:._-0/>
+<::._-0/>
+<_:._-0/>
+<A/>
+<_/>
+<:/>
+</doc>

test/oasis/p06fail1.xml

@@ -0,0 +1,13 @@
+<!--non-validating processors may pass this instance because they don't check the IDREFS attribute type-->
+<!DOCTYPE doc
+[
+<!ELEMENT doc (a|refs)*>
+<!ELEMENT a EMPTY>
+<!ELEMENT refs EMPTY>
+<!ATTLIST refs refs IDREFS #REQUIRED>
+<!ATTLIST a id ID #REQUIRED>
+]>
+<doc>
+<a id="A1"/><a id="A2"/><a id="A3"/>
+<refs refs=""/>
+</doc>

test/oasis/p06pass1.xml

@@ -0,0 +1,15 @@
+<!DOCTYPE doc
+[
+<!ELEMENT doc (a|refs)*>
+<!ELEMENT a EMPTY>
+<!ELEMENT refs EMPTY>
+<!ATTLIST refs refs IDREFS #REQUIRED>
+<!ATTLIST a id ID #REQUIRED>
+]>
+<doc>
+<a id="A1"/><a id="A2"/><a id="A3"/>
+<refs refs="A1 A2 A3"/>
+<refs refs="A1
+A2	A3"/>
+<refs refs="A1"/>
+</doc>

test/oasis/p07pass1.xml

@@ -0,0 +1,6 @@
+<!DOCTYPE doc
+[
+<!ELEMENT doc EMPTY>
+<!ATTLIST doc att (0|35a|A|-a|:a|a:|.|_a) #IMPLIED>
+]>
+<doc/>

test/oasis/p08fail1.xml

@@ -0,0 +1,10 @@
+<!--note: non-validating parsers may accept this document-->
+<!DOCTYPE doc
+[
+<!ELEMENT doc (A*)>
+<!ELEMENT A EMPTY>
+<!ATTLIST A att NMTOKENS #IMPLIED>
+]>
+<doc>
+<A att=""/>
+</doc>

test/oasis/p08fail2.xml

@@ -0,0 +1,10 @@
+<!--note: non-validating parsers may accept this document-->
+<!DOCTYPE doc
+[
+<!ELEMENT doc (A*)>
+<!ELEMENT A EMPTY>
+<!ATTLIST A att NMTOKENS #IMPLIED>
+]>
+<doc>
+<A att="abc / def"/>
+</doc>

test/oasis/p08pass1.xml

@@ -0,0 +1,12 @@
+<!DOCTYPE doc
+[
+<!ELEMENT doc (A*)>
+<!ELEMENT A EMPTY>
+<!ATTLIST A att NMTOKENS #IMPLIED>
+]>
+<doc>
+<A att="abc"/><A att="abc def . :"/><A att="
+abc
+def
+"/>
+</doc>

test/oasis/p09fail1.dtd

@@ -0,0 +1,2 @@
+<!ELEMENT doc EMPTY>
+<!ENTITY % ent1 "asdf%">

test/oasis/p09fail1.xml

@@ -0,0 +1,2 @@
+<!DOCTYPE doc SYSTEM "p09fail1.dtd">
+<doc/>

test/oasis/p09fail2.dtd

@@ -0,0 +1,2 @@
+<!ELEMENT doc EMPTY>
+<!ENTITY % ent1 "asdf&">

test/oasis/p09fail2.xml

@@ -0,0 +1,2 @@
+<!DOCTYPE doc SYSTEM "p09fail2.dtd">
+<doc/>

test/oasis/p09fail3.xml

@@ -0,0 +1,6 @@
+<!DOCTYPE doc
+[
+<!ELEMENT doc EMPTY>
+<!ENTITY % ent1 "asdf&#65">
+]>
+<doc/>

test/oasis/p09fail4.xml

@@ -0,0 +1,6 @@
+<!DOCTYPE doc
+[
+<!ELEMENT doc EMPTY>
+<!ENTITY % ent1 'a">
+]>
+<doc/>

test/oasis/p09fail5.xml

@@ -0,0 +1,6 @@
+<!DOCTYPE doc
+[
+<!ELEMENT doc EMPTY>
+<!ENTITY % ent1 "a'>
+]>
+<doc/>

test/oasis/p09pass1.dtd

@@ -0,0 +1,5 @@
+<!ELEMENT doc EMPTY>
+<!ENTITY % ent1 "">
+<!ENTITY ent2 "text2">
+<!ENTITY % ent3 "<!-- <!DOCTYPE <!ELEMENT <? '''&#34;&ent2; %ent1;">
+<!ENTITY % ent4 '""&#x27;&#39;"'>

test/oasis/p09pass1.xml

@@ -0,0 +1,2 @@
+<!DOCTYPE doc SYSTEM "p09pass1.dtd">
+<doc/>

test/oasis/p10fail1.xml

@@ -0,0 +1 @@
+<doc a="1 < 2"/>

test/oasis/p10fail2.xml

@@ -0,0 +1 @@
+<doc a="1 &"/>

test/oasis/p10fail3.xml

@@ -0,0 +1 @@
+<doc a='asd"/>

test/oasis/p10pass1.xml

@@ -0,0 +1,6 @@
+<doc>
+<A a="asdf>'&#34;>
+asdf
+	?>%"/>
+<A a='"">&#39;&#34;'/>
+</doc>