| 1 |
|
| 2 | module htm8
|
| 3 | {
|
| 4 |
|
| 5 | h8_id =
|
| 6 | Decl
|
| 7 |
|
| 8 | # CommentBegin, ProcessingBegin, CDataBegin are "pseudo-tokens", not visible
|
| 9 | | Comment | CommentBegin
|
| 10 | | Processing | ProcessingBegin
|
| 11 | | CData | CDataBegin
|
| 12 |
|
| 13 | | StartTag | StartEndTag | EndTag
|
| 14 |
|
| 15 | | DecChar | HexChar | CharEntity
|
| 16 |
|
| 17 | | RawData | HtmlCData
|
| 18 |
|
| 19 | | BadAmpersand | BadGreaterThan | BadLessThan
|
| 20 |
|
| 21 | | Invalid
|
| 22 | | EndOfStream
|
| 23 | generate [no_namespace_suffix] # cosmetic: call it h8_id, not h8_id_e
|
| 24 |
|
| 25 |
|
| 26 | h8_tag_id =
|
| 27 | TagName
|
| 28 | | AttrName
|
| 29 | | UnquotedValue | QuotedValue | MissingValue
|
| 30 | generate [no_namespace_suffix]
|
| 31 |
|
| 32 | attr_name =
|
| 33 | Ok # Found it
|
| 34 | | Eof
|
| 35 | | Error(int pos) # LexError
|
| 36 |
|
| 37 | attr_value =
|
| 38 | Missing # <a missing>
|
| 39 | | Empty # <a empty= >
|
| 40 | | Unquoted # <a unquoted=1 >
|
| 41 | | Quoted # <a quoted="1" >
|
| 42 | generate [no_namespace_suffix]
|
| 43 |
|
| 44 | # This API is maybe more natural, but has more allocations
|
| 45 | #
|
| 46 | # tag_lexer.Read()
|
| 47 |
|
| 48 | # # Unquoted, Quoted, Empty, Missing
|
| 49 | # (int tag_name_start, int tag_name_end, attr_value)
|
| 50 |
|
| 51 | # attr_value =
|
| 52 | # Missing # <a missing> - tag_name_end adds =""
|
| 53 |
|
| 54 | # | Empty (int equals_pos) # <a empty=>
|
| 55 |
|
| 56 | # # <a unquoted=foo>
|
| 57 | # # the first one has end_pos 0, and can be h8_id.ZeroPosition?
|
| 58 | # | Unquoted (List[Tuple[h8_id, end_pos]] tokens)
|
| 59 |
|
| 60 | # # <a quoted="foo">
|
| 61 | # | Quoted (List[Tuple[h8_id, end_pos]] tokens)
|
| 62 |
|
| 63 | # # Rather than raise an exception, we should have this for bad data
|
| 64 | # | Invalid(int pos)
|
| 65 |
|
| 66 |
|
| 67 | }
|
| 68 |
|