diff options
Diffstat (limited to 'addons/gdUnit4/src/update/GdMarkDownReader.gd')
| -rw-r--r-- | addons/gdUnit4/src/update/GdMarkDownReader.gd | 405 |
1 files changed, 405 insertions, 0 deletions
diff --git a/addons/gdUnit4/src/update/GdMarkDownReader.gd b/addons/gdUnit4/src/update/GdMarkDownReader.gd new file mode 100644 index 0000000..dab19e4 --- /dev/null +++ b/addons/gdUnit4/src/update/GdMarkDownReader.gd @@ -0,0 +1,405 @@ +@tool +extends RefCounted + +const GdUnitUpdateClient = preload("res://addons/gdUnit4/src/update/GdUnitUpdateClient.gd") + +const FONT_H1 := 22 +const FONT_H2 := 20 +const FONT_H3 := 18 +const FONT_H4 := 16 +const FONT_H5 := 14 +const FONT_H6 := 12 + +const HORIZONTAL_RULE := "[img=4000x2]res://addons/gdUnit4/src/update/assets/horizontal-line2.png[/img]" +const HEADER_RULE := "[font_size=%d]$1[/font_size]" +const HEADER_CENTERED_RULE := "[font_size=%d][center]$1[/center][/font_size]" + +const image_download_folder := "res://addons/gdUnit4/tmp-update/" + +const exclude_font_size := "\b(?!(?:(font_size))\b)" + +var md_replace_patterns := [ + # comments + [regex("(?m)^\\n?\\s*<!--[\\s\\S]*?-->\\s*\\n?"), ""], + + # horizontal rules + [regex("(?m)^[ ]{0,3}---$"), HORIZONTAL_RULE], + [regex("(?m)^[ ]{0,3}___$"), HORIZONTAL_RULE], + [regex("(?m)^[ ]{0,3}\\*\\*\\*$"), HORIZONTAL_RULE], + + # headers + [regex("(?m)^###### (.*)"), HEADER_RULE % FONT_H6], + [regex("(?m)^##### (.*)"), HEADER_RULE % FONT_H5], + [regex("(?m)^#### (.*)"), HEADER_RULE % FONT_H4], + [regex("(?m)^### (.*)"), HEADER_RULE % FONT_H3], + [regex("(?m)^## (.*)"), (HEADER_RULE + HORIZONTAL_RULE) % FONT_H2], + [regex("(?m)^# (.*)"), (HEADER_RULE + HORIZONTAL_RULE) % FONT_H1], + [regex("(?m)^(.+)=={2,}$"), HEADER_RULE % FONT_H1], + [regex("(?m)^(.+)--{2,}$"), HEADER_RULE % FONT_H2], + # html headers + [regex("<h1>((.*?\\R?)+)<\\/h1>"), (HEADER_RULE + HORIZONTAL_RULE) % FONT_H1], + [regex("<h1[ ]*align[ ]*=[ ]*\"center\">((.*?\\R?)+)<\\/h1>"), (HEADER_CENTERED_RULE + HORIZONTAL_RULE) % FONT_H1], + [regex("<h2>((.*?\\R?)+)<\\/h2>"), (HEADER_RULE + HORIZONTAL_RULE) % FONT_H2], + [regex("<h2[ ]*align[ ]*=[ ]*\"center\">((.*?\\R?)+)<\\/h2>"), (HEADER_CENTERED_RULE + HORIZONTAL_RULE) % FONT_H1], + [regex("<h3>((.*?\\R?)+)<\\/h3>"), HEADER_RULE % FONT_H3], + [regex("<h3[ ]*align[ ]*=[ ]*\"center\">((.*?\\R?)+)<\\/h3>"), HEADER_CENTERED_RULE % FONT_H3], + [regex("<h4>((.*?\\R?)+)<\\/h4>"), HEADER_RULE % FONT_H4], + [regex("<h4[ ]*align[ ]*=[ ]*\"center\">((.*?\\R?)+)<\\/h4>"), HEADER_CENTERED_RULE % FONT_H4], + [regex("<h5>((.*?\\R?)+)<\\/h5>"), HEADER_RULE % FONT_H5], + [regex("<h5[ ]*align[ ]*=[ ]*\"center\">((.*?\\R?)+)<\\/h5>"), HEADER_CENTERED_RULE % FONT_H5], + [regex("<h6>((.*?\\R?)+)<\\/h6>"), HEADER_RULE % FONT_H6], + [regex("<h6[ ]*align[ ]*=[ ]*\"center\">((.*?\\R?)+)<\\/h6>"), HEADER_CENTERED_RULE % FONT_H6], + + # asterics + #[regex("(\\*)"), "xxx$1xxx"], + + # extract/compile image references + [regex("!\\[(.*?)\\]\\[(.*?)\\]"), process_image_references], + # extract images with path and optional tool tip + [regex("!\\[(.*?)\\]\\((.*?)(( )+(.*?))?\\)"), process_image], + + # links + [regex("([!]|)\\[(.+)\\]\\(([^ ]+?)\\)"), "[url={\"url\":\"$3\"}]$2[/url]"], + # links with tool tip + [regex("([!]|)\\[(.+)\\]\\(([^ ]+?)( \"(.+)\")?\\)"), "[url={\"url\":\"$3\", \"tool_tip\":\"$5\"}]$2[/url]"], + # links to github, as shorted link + [regex("(https://github.*/?/(\\S+))"), '[url={"url":"$1", "tool_tip":"$1"}]#$2[/url]'], + + # embeded text + [regex("(?m)^[ ]{0,3}>(.*?)$"), "[img=50x14]res://addons/gdUnit4/src/update/assets/embedded.png[/img][i]$1[/i]"], + + # italic + bold font + [regex("[_]{3}(.*?)[_]{3}"), "[i][b]$1[/b][/i]"], + [regex("[\\*]{3}(.*?)[\\*]{3}"), "[i][b]$1[/b][/i]"], + # bold font + [regex("<b>(.*?)<\\/b>"), "[b]$1[/b]"], + [regex("[_]{2}(.*?)[_]{2}"), "[b]$1[/b]"], + [regex("[\\*]{2}(.*?)[\\*]{2}"), "[b]$1[/b]"], + # italic font + [regex("<i>(.*?)<\\/i>"), "[i]$1[/i]"], + [regex(exclude_font_size+"_(.*?)_"), "[i]$1[/i]"], + [regex("\\*(.*?)\\*"), "[i]$1[/i]"], + + # strikethrough font + [regex("<s>(.*?)</s>"), "[s]$1[/s]"], + [regex("~~(.*?)~~"), "[s]$1[/s]"], + [regex("~(.*?)~"), "[s]$1[/s]"], + + # handling lists + # using an image for dots + [regex("(?m)^[ ]{0,1}[*\\-+] (.*)$"), list_replace(0)], + [regex("(?m)^[ ]{2,3}[*\\-+] (.*)$"), list_replace(1)], + [regex("(?m)^[ ]{4,5}[*\\-+] (.*)$"), list_replace(2)], + [regex("(?m)^[ ]{6,7}[*\\-+] (.*)$"), list_replace(3)], + [regex("(?m)^[ ]{8,9}[*\\-+] (.*)$"), list_replace(4)], + + # code + [regex("``([\\s\\S]*?)``"), code_block("$1")], + [regex("`([\\s\\S]*?)`{1,2}"), code_block("$1")], +] + +var code_block_patterns := [ + # code blocks, code blocks looks not like code blocks in richtext + [regex("```(javascript|python|shell|gdscript|gd)([\\s\\S]*?\n)```"), code_block("$2", true)], +] + +var _img_replace_regex := RegEx.new() +var _image_urls := PackedStringArray() +var _on_table_tag := false +var _client: GdUnitUpdateClient + + +static func regex(pattern: String) -> RegEx: + var regex_ := RegEx.new() + var err := regex_.compile(pattern) + if err != OK: + push_error("error '%s' checked pattern '%s'" % [err, pattern]) + return null + return regex_ + + +func _init() -> void: + @warning_ignore("return_value_discarded") + _img_replace_regex.compile("\\[img\\]((.*?))\\[/img\\]") + + +func set_http_client(client: GdUnitUpdateClient) -> void: + _client = client + + +@warning_ignore("return_value_discarded") +func _notification(what: int) -> void: + if what == NOTIFICATION_PREDELETE: + # finally remove_at the downloaded images + for image in _image_urls: + DirAccess.remove_absolute(image) + DirAccess.remove_absolute(image + ".import") + + +func list_replace(indent: int) -> String: + var replace_pattern := "[img=12x12]res://addons/gdUnit4/src/update/assets/dot2.png[/img]" if indent %2 else "[img=12x12]res://addons/gdUnit4/src/update/assets/dot1.png[/img]" + replace_pattern += " $1" + + for index in indent: + replace_pattern = replace_pattern.insert(0, " ") + return replace_pattern + + +func code_block(replace: String, border: bool = false) -> String: + if border: + return """ + [img=1400x14]res://addons/gdUnit4/src/update/assets/border_top.png[/img] + [indent][color=GRAY][font_size=16]%s[/font_size][/color][/indent] + [img=1400x14]res://addons/gdUnit4/src/update/assets/border_bottom.png[/img] + """.dedent() % replace + return "[code][bgcolor=DARK_SLATE_GRAY][color=GRAY][font_size=16]%s[/font_size][/color][/bgcolor][/code]" % replace + + +func convert_text(input: String) -> String: + input = process_tables(input) + + for pattern: Array in md_replace_patterns: + var regex_: RegEx = pattern[0] + var bb_replace: Variant = pattern[1] + if bb_replace is Callable: + @warning_ignore("unsafe_method_access") + input = await bb_replace.call(regex_, input) + else: + @warning_ignore("unsafe_cast") + input = regex_.sub(input, bb_replace as String, true) + return input + + +func convert_code_block(input: String) -> String: + for pattern: Array in code_block_patterns: + var regex_: RegEx = pattern[0] + var bb_replace: Variant = pattern[1] + if bb_replace is Callable: + @warning_ignore("unsafe_method_access") + input = await bb_replace.call(regex_, input) + else: + @warning_ignore("unsafe_cast") + input = regex_.sub(input, bb_replace as String, true) + return input + + +func to_bbcode(input: String) -> String: + var re := regex("(?m)```[\\s\\S]*?```") + var current_pos := 0 + var as_bbcode := "" + + # we split by code blocks to handle this blocks customized + for result in re.search_all(input): + # Add text before code block + if result.get_start() > current_pos: + as_bbcode += await convert_text(input.substr(current_pos, result.get_start() - current_pos)) + # Add code block + as_bbcode += await convert_code_block(result.get_string()) + current_pos = result.get_end() + + # Add remaining text after last code block + if current_pos < input.length(): + as_bbcode += await convert_text(input.substr(current_pos)) + return as_bbcode + + +func process_tables(input: String) -> String: + var bbcode := PackedStringArray() + var lines: Array[String] = Array(input.split("\n") as Array, TYPE_STRING, "", null) + while not lines.is_empty(): + if is_table(lines[0]): + bbcode.append_array(parse_table(lines)) + continue + @warning_ignore("return_value_discarded", "unsafe_cast") + bbcode.append(lines.pop_front() as String) + return "\n".join(bbcode) + + +class GdUnitMDReaderTable: + var _columns: int + var _rows: Array[Row] = [] + + class Row: + var _cells := PackedStringArray() + + + func _init(cells: PackedStringArray, columns: int) -> void: + _cells = cells + for i in range(_cells.size(), columns): + @warning_ignore("return_value_discarded") + _cells.append("") + + + func to_bbcode(cell_sizes: PackedInt32Array, bold: bool) -> String: + var cells := PackedStringArray() + for cell_index in _cells.size(): + var cell: String = _cells[cell_index] + if cell.strip_edges() == "--": + cell = create_line(cell_sizes[cell_index]) + if bold: + cell = "[b]%s[/b]" % cell + @warning_ignore("return_value_discarded") + cells.append("[cell]%s[/cell]" % cell) + return "|".join(cells) + + + func create_line(length: int) -> String: + var line := "" + for i in length: + line += "-" + return line + + + func _init(columns: int) -> void: + _columns = columns + + + func parse_row(line :String) -> bool: + # is line containing cells? + if line.find("|") == -1: + return false + _rows.append(Row.new(line.split("|"), _columns)) + return true + + + func calculate_max_cell_sizes() -> PackedInt32Array: + var cells_size := PackedInt32Array() + for column in _columns: + @warning_ignore("return_value_discarded") + cells_size.append(0) + + for row_index in _rows.size(): + var row: Row = _rows[row_index] + for cell_index in row._cells.size(): + var cell_size: int = cells_size[cell_index] + var size := row._cells[cell_index].length() + if size > cell_size: + cells_size[cell_index] = size + return cells_size + + + @warning_ignore("return_value_discarded") + func to_bbcode() -> PackedStringArray: + var cell_sizes := calculate_max_cell_sizes() + var bb_code := PackedStringArray() + + bb_code.append("[table=%d]" % _columns) + for row_index in _rows.size(): + bb_code.append(_rows[row_index].to_bbcode(cell_sizes, row_index==0)) + bb_code.append("[/table]\n") + return bb_code + + +func parse_table(lines: Array) -> PackedStringArray: + var line: String = lines[0] + var table := GdUnitMDReaderTable.new(line.count("|") + 1) + while not lines.is_empty(): + line = lines.pop_front() + if not table.parse_row(line): + break + return table.to_bbcode() + + +func is_table(line: String) -> bool: + return line.find("|") != -1 + + +func open_table(line: String) -> String: + _on_table_tag = true + return "[table=%d]" % (line.count("|") + 1) + + +func close_table() -> String: + _on_table_tag = false + return "[/table]" + + +func extract_cells(line: String, bold := false) -> String: + var cells := "" + for cell in line.split("|"): + if bold: + cell = "[b]%s[/b]" % cell + cells += "[cell]%s[/cell]" % cell + return cells + + +func process_image_references(p_regex: RegEx, p_input: String) -> String: + #return p_input + + # exists references? + var matches := p_regex.search_all(p_input) + if matches.is_empty(): + return p_input + # collect image references and remove_at it + var references := Dictionary() + var link_regex := regex("\\[(\\S+)\\]:(\\S+)([ ]\"(.*)\")?") + # create copy of original source to replace checked it + var input := p_input.replace("\r", "") + var extracted_references := p_input.replace("\r", "") + for reg_match in link_regex.search_all(input): + var line := reg_match.get_string(0) + "\n" + var ref := reg_match.get_string(1) + #var topl_tip = reg_match.get_string(4) + # collect reference and url + references[ref] = reg_match.get_string(2) + extracted_references = extracted_references.replace(line, "") + + # replace image references by collected url's + for reference_key: String in references.keys(): + var regex_key := regex("\\](\\[%s\\])" % reference_key) + for reg_match in regex_key.search_all(extracted_references): + var ref: String = reg_match.get_string(0) + var image_url: String = "](%s)" % references.get(reference_key) + extracted_references = extracted_references.replace(ref, image_url) + return extracted_references + + +@warning_ignore("return_value_discarded") +func process_image(p_regex: RegEx, p_input: String) -> String: + #return p_input + var to_replace := PackedStringArray() + var tool_tips := PackedStringArray() + # find all matches + var matches := p_regex.search_all(p_input) + if matches.is_empty(): + return p_input + for reg_match in matches: + # grap the parts to replace and store temporay because a direct replace will distort the offsets + to_replace.append(p_input.substr(reg_match.get_start(0), reg_match.get_end(0))) + # grap optional tool tips + tool_tips.append(reg_match.get_string(5)) + # finally replace all findings + for replace in to_replace: + var re := p_regex.sub(replace, "[img]$2[/img]") + p_input = p_input.replace(replace, re) + return await _process_external_image_resources(p_input) + + +func _process_external_image_resources(input: String) -> String: + @warning_ignore("return_value_discarded") + DirAccess.make_dir_recursive_absolute(image_download_folder) + # scan all img for external resources and download it + for value in _img_replace_regex.search_all(input): + if value.get_group_count() >= 1: + var image_url: String = value.get_string(1) + # if not a local resource we need to download it + if image_url.begins_with("http"): + if OS.is_stdout_verbose(): + prints("download image:", image_url) + var response := await _client.request_image(image_url) + if response.status() == 200: + var image := Image.new() + var error := image.load_png_from_buffer(response.get_body()) + if error != OK: + prints("Error creating image from response", error) + # replace characters where format characters + var new_url := image_download_folder + image_url.get_file().replace("_", "-") + if new_url.get_extension() != 'png': + new_url = new_url + '.png' + var err := image.save_png(new_url) + if err: + push_error("Can't save image to '%s'. Error: %s" % [new_url, error_string(err)]) + @warning_ignore("return_value_discarded") + _image_urls.append(new_url) + input = input.replace(image_url, new_url) + return input |