chore: manpage

README.md

@@ -34,9 +34,7 @@ Supported combinators are all the "basic" ones:
 
 ### TODO
 
-- [ ] `--text` option to only get the text in the matched elements
 - [ ] Universal selector (`*` to match any element)
-- [ ] Attribute selectors (`[attr="value"]`)
 - [ ] A way to "group" selectors, e.g. `aside {p, footer}` to select all `p`s and `footer`s in `aside`s ? 
 
 ## Usage
@@ -57,6 +55,8 @@ Where:
 *   `-1`, `--first-only`: Return only the first match
 *   `-e`, `--errors`: print warnings
 *   `-t`, `--text`: Print only the [innerText](https://developer.mozilla.org/fr/docs/Web/API/HTMLElement/innerText) of the matched elements
+*   `-t`, `--text`: Print only the [innerText](https://developer.mozilla.org/fr/
+*   `-a`, `--select-attribute`: Print the value of the attribute on matched elements. Supersedes -t.
 
 ## Motivation
 

html.lua

@@ -159,7 +159,7 @@ function M.make_dom_element( tag_name, parent_elem )
 			end
 
 			return text
-		end
+		end,
 	}
 
 	if parent_elem then
@@ -207,76 +207,27 @@ function M.tokenise( content )
 	local currently_opened_quotes = nil
 	local text_memory = ""
 
-	local skipping_from = nil
-	local skip_target = nil
-	local skip_mode = "before"
-
-	local function set_skipping_to( str, mode )
-		mode = mode or "before"
-		if mode ~= "before" and mode ~= "after" then
-			error("Unexpected skipping mode: " .. mode .. ", in looking for " .. str)
-		end
-
-		skip_target = str
-		skip_mode = mode
-	end
-
-
-
 	local i = 1
 
 	while i <= #content do
 		local char = content:sub(i,i)
 
-		if skip_target ~= nil then
-			if skipping_from == nil then
-				skipping_from = i
-			end
-
-			if skip_mode == "before" then
-				local end_i =  i + #skip_target - 1
-
-				if trim(content:sub(i, end_i)) == skip_target then
-					table.insert( TOKENS, {type="TEXT", value=content:sub(skipping_from, i-1)} )
-
-					-- release from skip
-					--i = end_i - 1
-					i = i - 1
-					skip_target = nil
-					skipping_from = nil
-				end
-
-				goto continue
-			else
-				local start_i =  i - #skip_target + 1
-
-				if trim(content:sub(start_i, i)) == skip_target then
-					table.insert( TOKENS, {type="TEXT", value=content:sub(skipping_from, start_i-1)} )
-
-					-- release from skip
-					i = start_i
-					skip_target = nil
-					skipping_from = nil
-				end
-
-				goto continue
-			end
-
-
-
-		end
-
-
 
 
 		if char == "<" then
 			if content:sub(i, i+3) == "<!--" then
-				set_skipping_to("-->", "after")
+				local end_i = content:find("-->", i+3, true)
+				if end_i then
+					i = end_i + 2
+				else
+					i = #content
+				end
+
 				goto continue
 			end
 
 			if content:sub(i, i+1) == "<!" then
-				i = content:find(">", i)
+				i = content:find(">", i, true)
 				goto continue
 			end
 
@@ -308,7 +259,7 @@ function M.tokenise( content )
 						if RAW_TEXT_TAGS[word] then
 							logger.printerr("Warning: "..word.." tags may contain text that would be incorrectly parsed as HTML.")
 							-- made possible because of the whitespace removal at the start
-							set_skipping_to("</" .. word)
+							i = content:find("</"..word, i, true) - 1
 						end
 					end
 
@@ -338,12 +289,12 @@ function M.tokenise( content )
 					if TOKENS[#TOKENS] and ( TOKENS[#TOKENS].type == "START_OPENING_TAG" ) then
 						if RAW_TEXT_TAGS[word] then
 							logger.printerr("Warning: "..word.." tags may contain text that would be incorrectly parsed as HTML.")
-							-- made possible because of the whitespace removal at the start
-							set_skipping_to("</" .. word)
 							text_memory = ""
 
 							-- advance to closing ">"
-							i = content:find(">", i)
+							i = content:find(">", i, true)
+							-- made possible because of the whitespace removal at the start
+							i = content:find("</"..word, i, true) - 1
 						end
 					end
 
@@ -356,16 +307,13 @@ function M.tokenise( content )
 				end
 			end
 
-
 			if char == "'" or char == '"' then
-				if currently_opened_quotes == char then
+				-- found matching closing quote type
+				if char == currently_opened_quotes then
 					currently_opened_quotes = nil
-				else
+				elseif currently_opened_quotes == nil then
 					currently_opened_quotes = char
 				end
-
-				text_memory = text_memory .. char
-				goto continue
 			end
 
 			text_memory = text_memory .. char
@@ -502,7 +450,7 @@ function M.parse_tokens_into_document( TOKENS )
 
 
 			if in_opening_tag_for then
-				local pattern = "([%w-]+)=['\"](.-)['\"]"
+				local pattern = "([%w-]+)=['\"](.+)['\"]"
 
 				local name, raw_value = token.value:match(pattern)
 

htmlq.1

@@ -17,6 +17,9 @@ Print warning messages
 .TP
 .BR \-t ", " \-\-text
 Print only the innerText of matched elements
+.TP
+.BR \-a ", " \-\-select\-attribute
+Print the value of the attribute on matched elements. Supersedes -t.
 .SH ARGUMENTS
 .TP
 .I html_path_or_minus

main.lua

@@ -42,6 +42,7 @@ local function print_usage()
 	logger.print("  -1, --first-only: return only the first match")
 	logger.print("  -e, --errors: print warnings")
 	logger.print("  -t, --text: Print only the innerText of the matched elements")
+	logger.print("  -a, --select-attribute: Print the value of the attribute on matched elements. Supersedes -t.")
 end
 
 
@@ -59,18 +60,26 @@ local FLAGS = {
 	FIRST_ONLY = {},
 	DO_PRINT_ERRORS = {},
 	INNER_TEXT = {},
+	SELECT_ATTRIBUTE = {}
 }
 
 local LONGHAND_FLAGS = {
 	["first-only"] = FLAGS.FIRST_ONLY,
 	["errors"] = FLAGS.DO_PRINT_ERRORS,
 	["text"] = FLAGS.INNER_TEXT,
+	["select-attribute"] = FLAGS.SELECT_ATTRIBUTE,
 }
 
 local SHORTHAND_FLAGS = {
 	["1"] = FLAGS.FIRST_ONLY,
 	["e"] = FLAGS.DO_PRINT_ERRORS,
 	["t"] = FLAGS.INNER_TEXT,
+	["a"] = FLAGS.SELECT_ATTRIBUTE,
+}
+
+
+local FLAG_NEEDS_VALUE = {
+	[FLAGS.SELECT_ATTRIBUTE] = true,
 }
 
 
@@ -84,44 +93,103 @@ end
 local flags = {}
 local positionals = {}
 
-for _, argument in ipairs(arg) do
+local i = 1
+while i <= #arg do
+	local argument = arg[i]
+
+	-- Handle shorthand flags (-a, -1, etc.)
 	if argument:match("^%-%w+$") then
-		for letter in argument:sub(2):gmatch("(%w)") do
-			if not SHORTHAND_FLAGS[letter] then
-				logger.printerr("Unknown flag: -"..letter..".")
+		local flag_str = argument:sub(2)
+
+		-- Handle single-letter flags
+		if #flag_str == 1 then
+			local letter = flag_str
+			local flag = SHORTHAND_FLAGS[letter]
+
+			if not flag then
+				logger.printerr("Unknown flag: -"..letter)
 				print_usage()
 				os.exit(RETURN_CODES.ARGUMENTS_ERROR)
 			end
 
+			-- Handle flags that require values
+			if FLAG_NEEDS_VALUE[flag] then
+				if i == #arg then
+					logger.printerr("Flag -"..letter.." requires a value")
+					os.exit(RETURN_CODES.ARGUMENTS_ERROR)
+				end
+				flags[flag] = arg[i+1]
+				i = i + 2  -- Skip next argument as it's the value
+			else
+				-- Handle regular boolean flags
+				if flags[flag] then
+					logger.printerr("Warning: passed -"..letter.." flag already!")
+				end
+				flags[flag] = true
+				i = i + 1
+			end
+
+		else
+			-- Handle grouped flags (-abc)
+			for letter in flag_str:gmatch("(%w)") do
 				local flag = SHORTHAND_FLAGS[letter]
 
+				if not flag then
+					logger.printerr("Unknown flag in group: -"..letter)
+					print_usage()
+					os.exit(RETURN_CODES.ARGUMENTS_ERROR)
+				end
+
+				if FLAG_NEEDS_VALUE[flag] then
+					logger.printerr("Cannot use value-taking flags in groups: -"..letter)
+					os.exit(RETURN_CODES.ARGUMENTS_ERROR)
+				end
+
 				if flags[flag] then
 					logger.printerr("Warning: passed -"..letter.." flag already!")
 				end
-
 				flags[flag] = true
 			end
-	elseif argument:match("^%-%-[%w%-]+$") then
+			i = i + 1
+		end
+
+		-- Handle long flags (--flag)
+	elseif argument:match("^%-%-") then
 		local flagname = argument:sub(3)
-		if not LONGHAND_FLAGS[flagname] then
-			logger.printerr("Unknown flag: --"..flagname..".")
+		local flag = LONGHAND_FLAGS[flagname]
+
+		if not flag then
+			logger.printerr("Unknown flag: --"..flagname)
 			print_usage()
 			os.exit(RETURN_CODES.ARGUMENTS_ERROR)
 		end
 
-		local flag = LONGHAND_FLAGS[flagname]
-
+		-- Handle flags that require values
+		if FLAG_NEEDS_VALUE[flag] then
+			if i == #arg then
+				logger.printerr("Flag --"..flagname.." requires a value")
+				os.exit(RETURN_CODES.ARGUMENTS_ERROR)
+			end
+			flags[flag] = arg[i+1]
+			i = i + 2  -- Skip next argument as it's the value
+		else
+			-- Handle regular boolean flags
 			if flags[flag] then
 				logger.printerr("Warning: passed --"..flagname.." flag already!")
 			end
-
 			flags[flag] = true
+			i = i + 1
+		end
+
 	else
+		-- Handle positional arguments
 		table.insert(positionals, argument)
+		i = i + 1
 	end
 end
 
 
+
 if flags[ FLAGS.DO_PRINT_ERRORS ] then
 	logger.enable_printing_errors()
 end
@@ -238,24 +306,51 @@ if #elements == 0 then
 	os.exit( RETURN_CODES.NOTHING_FOUND )
 end
 
-
+local MAX_NUMBER_OF_ELEMENTS_TO_SHOW = #elements
 if flags[FLAGS.FIRST_ONLY] then
-	if #elements > 0 then
-		if flags[FLAGS.INNER_TEXT] then
-			logger.print( elements[1]:inner_text() )
-			os.exit( RETURN_CODES.OK )
+	MAX_NUMBER_OF_ELEMENTS_TO_SHOW = 1
 end
 
-		logger.print( HTML.tostring( elements[1] ) )
+
+
+
+
+local attr = flags[FLAGS.SELECT_ATTRIBUTE]
+if attr then
+	local spoof_nil = {}
+	local attrs = {}
+
+	local i = 1
+	while i <= MAX_NUMBER_OF_ELEMENTS_TO_SHOW do
+		local el = elements[i]
+
+		local attribute_value = el.attributes[attr]
+
+		table.insert( attrs, attribute_value or spoof_nil )
+
+		i = i+1
 	end
 
-	os.exit( RETURN_CODES.OK )
+	local nb_non_nil_values = 0
+	for _, val in ipairs(attrs) do
+		if val ~= spoof_nil then
+			nb_non_nil_values = nb_non_nil_values + 1
+		end
 	end
 
-for _, el in ipairs(elements) do
-		if flags[FLAGS.INNER_TEXT] then
-			logger.print( el:inner_text() )
+	if nb_non_nil_values == 0 then
+		os.exit( RETURN_CODES.NOTHING_FOUND )
+	end
+
+	for _, val in ipairs(attrs) do
+		if val ~= spoof_nil then
+			print(val)
 		else
-			logger.print( HTML.tostring(el) )
+			print()
 		end
 	end
+
+	os.exit( RETURN_CODES.OK )
+end
+
+end