dymokomi/luce-prism
Source of release 0.6.0
0.6.0 / src / codecs / markdown_syntax.lucb
## Line and block recognizers shared by the Markdown reader and writer.
from storage import Buffer, List, close_names, text_copy
from codecs.common import trim, find
pub func lines(source: str) -> List[str]!:
var output: List[str]
errdefer output.close()
var start: usize = 0
for i in 0..=source.length:
if i == source.length or source.bytes[i] == '\n':
let end = i - 1 if i > start and source.bytes[i - 1] == '\r' else i
try output.push((str)source.bytes[start..<end])
start = i + 1
if output.count > 0 and output.items[output.count - 1].length == 0:
output.count -= 1
return output
pub func rstrip(text: str) -> str:
var end = text.length
while end > 0 and (text.bytes[end - 1] == ' ' or text.bytes[end - 1] == '\t' or text.bytes[end - 1] == '\r'):
end -= 1
return (str)text.bytes[..<end]
pub func indent(text: str) -> usize:
var i: usize = 0
while i < text.length and (text.bytes[i] == ' ' or text.bytes[i] == '\t'):
i += 1
return i
pub func blank(text: str) -> bool:
return indent(text) == text.length
pub func heading(text: str) -> usize:
let i = indent(text)
var count: usize = 0
while i + count < text.length and text.bytes[i + count] == '#':
count += 1
return count if count > 0 and count <= 6 and i + count < text.length and text.bytes[i + count] == ' ' else 0
pub func thematic(text: str) -> bool:
var marker: u8 = 0
var count: usize = 0
for byte in text.bytes:
if byte == ' ' or byte == '\t':
continue
if marker == 0:
marker = byte
if byte != marker or (byte != '-' and byte != '*' and byte != '_'):
return false
count += 1
return count >= 3
pub struct Fence:
pub var marker: u8
pub var length: usize
pub var language: str
pub func fence(text: str) -> Fence?:
let start = indent(text)
if start == text.length or (text.bytes[start] != '`' and text.bytes[start] != '~'):
return none
let marker = text.bytes[start]
var end = start
while end < text.length and text.bytes[end] == marker:
end += 1
if end - start < 3:
return none
let rest = rstrip((str)text.bytes[end..])
var first: usize = 0
while first < rest.length and rest.bytes[first] == ' ':
first += 1
return Fence(marker = marker, length = end - start, language = (str)rest.bytes[first..])
pub func closes(text: str, opening: Fence) -> bool:
let start = indent(text)
var end = start
while end < text.length and text.bytes[end] == opening.marker:
end += 1
return end - start >= opening.length and rstrip((str)text.bytes[end..]).length == 0
pub func quote(text: str) -> bool:
let start = indent(text)
return start < text.length and text.bytes[start] == '>'
pub struct Marker:
pub var ordered: bool
pub var column: usize
pub func list_marker(text: str) -> Marker?:
let start = indent(text)
if start == text.length:
return none
let byte = text.bytes[start]
if byte == '-' or byte == '*' or byte == '+':
if start + 1 < text.length and text.bytes[start + 1] == ' ':
return Marker(column = start + 2)
return none
var end = start
while end < text.length and text.bytes[end] >= '0' and text.bytes[end] <= '9':
end += 1
if end > start and end + 1 < text.length and (text.bytes[end] == '.' or text.bytes[end] == ')') and text.bytes[end + 1] == ' ':
return Marker(ordered = true, column = end + 2)
return none
pub func block_start(text: str) -> bool:
return heading(text) > 0 or thematic(text) or fence(text) != none or quote(text) or list_marker(text) != none
pub func table_separator(text: str) -> bool:
let trimmed = trim(text)
if find(trimmed, "|") == none:
return false
var dash = false
for byte in trimmed.bytes:
if byte == '-':
dash = true
elif byte != '|' and byte != ':' and byte != ' ' and byte != '\t':
return false
return dash
pub func table_start(text: const str[], index: usize) -> bool:
return index + 1 < text.length and not blank(text[index]) and find(text[index], "|") != none and not table_separator(text[index]) and table_separator(text[index + 1])
pub func table_cells(text: str) -> List[Buffer]!:
let trimmed = trim(text)
var output: List[Buffer]
errdefer close_names(&output)
var cell: Buffer
defer cell.close()
var i: usize = 0
while i <= trimmed.length:
if i < trimmed.length and trimmed.bytes[i] == '\\' and i + 1 < trimmed.length and trimmed.bytes[i + 1] == '|':
try cell.byte('|')
i += 2
continue
if i == trimmed.length or trimmed.bytes[i] == '|':
var value = try text_copy(trim(cell.view()))
errdefer value.close()
try output.push(value)
cell.used = 0
else:
try cell.byte(trimmed.bytes[i])
i += 1
if output.count > 0 and output.items[0].used == 0:
output.items[0].close()
output.remove(0)
if output.count > 0 and output.items[output.count - 1].used == 0:
output.items[output.count - 1].close()
output.count -= 1
return output