.gitignore

@@ -2,5 +2,3 @@
 *.luastatic.c
 # Compiled executable from main.lua
 main
-# Same but with correct name
-htmlq

README.md

@@ -41,23 +41,16 @@ Supported combinators are all the "basic" ones:
 
 ## Usage
 
-Once compiled, you can run Htmlq using the following command:
-
 ```
-./htmlq [FLAGS] <html_path_or_minus> <css_selector>
+Usage: lua main.lua [FLAGS] <html_path_or_minus> <css_selector>
+  html_path_or_minus: Path to HTML file or '-' for stdin
+  css_selector: CSS selector to search for
+
+  Flags:
+  -f, --first-only: return only the first match
+  -q, --quiet: Don't print warnings
 ```
 
-Where:
-
-*   `<html_path_or_minus>` is the path to the HTML file you want to parse, or `-` to read from stdin.
-*   `<css_selector>` is the CSS selector you want to use to query the HTML.
-
-### Flags
-
-*   `-1`, `--first-only`: Return only the first match
-*   `-q`, `--quiet`: Don't print warnings
-*   `-t`, `--text`: Print only the [innerText](https://developer.mozilla.org/fr/docs/Web/API/HTMLElement/innerText) of the matched elements
-
 ## Motivation
 
 I needed this for a specific need of mine, where I wanted to systematically extract the HTML starting with an element with a certain id, up to the closing tag. While I could probably have hacked something together for this one-time use case, in typical programmer spirit, I decided to create a tool.
@@ -85,7 +78,25 @@ luarocks install luastatic
 Once `luastatic` is installed, you can compile Htmlq by running the following command in your terminal, from the project's root directory:
 
 ```
-luastatic main.lua css.lua html.lua logging.lua /usr/lib/liblua5.4.so -o htmlq
+luastatic main.lua css.lua html.lua logging.lua /usr/lib/liblua5.4.so
 ```
 
 Note that all `.lua` files from the project need to be specified, with `main.lua` as the first one. Also, the path to `liblua` may vary according to your system. The example provided is for an installation on EndeavourOS.
+
+## Running
+
+Once compiled, you can run Htmlq using the following command:
+
+```
+./htmlq [FLAGS] <html_path_or_minus> <css_selector>
+```
+
+Where:
+
+*   `<html_path_or_minus>` is the path to the HTML file you want to parse, or `-` to read from stdin.
+*   `<css_selector>` is the CSS selector you want to use to query the HTML.
+
+### Flags
+
+*   `-f`, `--first-only`: Return only the first match
+*   `-q`, `--quiet`: Don't print warnings

css.lua

@@ -66,8 +66,7 @@ local function parse_compound_selector( tokeniser )
 		tag_name = nil,
 		id = nil,
 		class = {},
-		attributes_values = {},
-		attributes_present = {},
+		attributes = {},
 	}
 
 	--local selectors = {}
@@ -107,37 +106,6 @@ local function parse_compound_selector( tokeniser )
 			end
 			--table.insert(selectors, {type = "id", value = name})
 			selector.id = name
-		elseif char == "[" then
-			tokeniser.next() -- consume leading [
-
-			local name = tokeniser.read_identifier()
-
-			if tokeniser.peek() == "=" then
-				tokeniser.next()
-
-				if tokeniser.peek() ~= "\"" then
-					error("Expected opening quote \" at pos " .. tokeniser.pos() )
-				end
-				tokeniser.next() -- consume leading "
-
-				local value = ""
-				while tokeniser.peek() ~= "\"" do
-					value = value .. tokeniser.peek()
-					tokeniser.next()
-				end
-
-				tokeniser.next() -- consume trailing "
-
-				selector.attributes_values[name] = value
-			else
-				table.insert( selector.attributes_present, name )
-			end
-
-			if tokeniser.peek() ~= "]" then
-				error("Expected closing bracket (']') at " .. tokeniser.pos())
-			end
-
-			tokeniser.next() -- consume trailing ]
 		else
 			break
 		end

html.lua

@@ -5,11 +5,11 @@ local function trim(str)
 end
 
 local function shallow_copy(t)
-	local t2 = {}
-	for k,v in pairs(t) do
-		t2[k] = v
-	end
-	return t2
+  local t2 = {}
+  for k,v in pairs(t) do
+    t2[k] = v
+  end
+  return t2
 end
 
 
@@ -39,62 +39,6 @@ local VOID_TAGS = {
 	wbr = true,
 }
 
-local INLINE_TAGS = {
-	-- Text formatting
-	a = true,
-	abbr = true,
-	b = true,
-	bdi = true,
-	bdo = true,
-	cite = true,
-	code = true,
-	data = true,
-	dfn = true,
-	em = true,
-	i = true,
-	kbd = true,
-	mark = true,
-	q = true,
-	ruby = true,
-	s = true,
-	samp = true,
-	small = true,
-	span = true,
-	strong = true,
-	sub = true,
-	sup = true,
-	time = true,
-	u = true,
-	var = true,
-
-	-- Interactive elements
-	button = true,
-	label = true,
-	select = true,
-	textarea = true,
-
-	-- Media/content
-	img = true,
-	picture = true,
-	map = true,
-	object = true,
-
-	-- Line break
-	br = true,
-	wbr = true,
-
-	-- Forms
-	input = true,
-	output = true,
-	progress = true,
-	meter = true,
-
-	-- Scripting
-	script = true,
-	noscript = true,
-	template = true,
-}
-
 
 function M.make_dom_element( tag_name, parent_elem )
 	local o = {
@@ -142,23 +86,6 @@ function M.make_dom_element( tag_name, parent_elem )
 			for _, child in ipairs(self.children or {}) do
 				child:foreach( fn )
 			end
-		end,
-
-		inner_text = function(self)
-			if self.tag_name == ":text" then
-				return self.content
-			end
-
-			local text = ""
-			for _, child in ipairs(self.children) do
-				text = text .. child:inner_text()
-
-				if not INLINE_TAGS[child.tag_name] then
-					text = text .. "\n"
-				end
-			end
-
-			return text
 		end
 	}
 
@@ -386,59 +313,45 @@ end
 
 
 function M.check_simple_selector(element, selector)
-	-- Skip text nodes
-	if element.tag_name == ":text" then
-		return false
-	end
-
-	-- Check tag name if specified
-	if selector.tag_name and element.tag_name ~= selector.tag_name then
-		return false
-	end
-
-	-- Check ID if specified
-	if selector.id and element.attributes.id ~= selector.id then
-		return false
-	end
-
-	-- Check classes if specified
-	if selector.class and #selector.class > 0 then
-		local element_classes = element.attributes.class
-		if not element_classes then
+		-- Skip text nodes
+		if element.tag_name == ":text" then
 			return false
 		end
 
-		for _, class in ipairs(selector.class) do
-			local found = false
-			for _, elem_class in ipairs(element_classes) do
-				if elem_class == class then
-					found = true
-					break
-				end
-			end
-			if not found then
+		-- Check tag name if specified
+		if selector.tag_name and element.tag_name ~= selector.tag_name then
+			return false
+		end
+
+		-- Check ID if specified
+		if selector.id and element.attributes.id ~= selector.id then
+			return false
+		end
+
+		-- Check classes if specified
+		if selector.class and #selector.class > 0 then
+			local element_classes = element.attributes.class
+			if not element_classes then
 				return false
 			end
-		end
-	end
 
-	for attr_name, attr_value in pairs(selector.attributes_values) do
-		local elem_attr_value = element.attributes[attr_name]
-		if elem_attr_value ~= attr_value then
-			return false
+			for _, class in ipairs(selector.class) do
+				local found = false
+				for _, elem_class in ipairs(element_classes) do
+					if elem_class == class then
+						found = true
+						break
+					end
+				end
+				if not found then
+					return false
+				end
+			end
 		end
-	end
 
-	-- Check attribute presence selectors
-	for _, attr_name in ipairs(selector.attributes_present) do
-		if not element.attributes[attr_name] then
-			return false
-		end
+		return true
 	end
 
-	return true
-end
-
 function M.query_simple_selector(document, selector)
 	local matches = {}
 
@@ -611,7 +524,7 @@ function M.clean_text_nodes(node)
 		return
 	end
 
-	node.content = node.content:gsub("%s+", " ")
+	node.content = trim( node.content:gsub("%s+", " ") )
 end
 
 
@@ -646,7 +559,7 @@ function M._tostring(node, indent, include_internal_pseudoelements)
 	local node_name = ""
 
 	if not is_pseudo_element or include_internal_pseudoelements then
-		-- Print the current node's tag name
+	-- Print the current node's tag name
 		node_name = node_name .. "\n" .. indent_str .. "<" .. (node.tag_name or ":root")
 	end
 

main.lua

@@ -21,9 +21,8 @@ local function print_usage()
     logger.print("  css_selector: CSS selector to search for")
 		logger.print()
 		logger.print("  Flags:")
-    logger.print("  -1, --first-only: return only the first match")
+    logger.print("  -f, --first-only: return only the first match")
     logger.print("  -q, --quiet: Don't print warnings")
-    logger.print("  -t, --text: Print only the innerText of the matched elements")
     os.exit(1)
 end
 
@@ -33,20 +32,17 @@ end
 
 local FLAGS = {
 	FIRST_ONLY = {},
-	DO_PRINT_ERRORS = {},
-	INNER_TEXT = {},
+	NO_PRINT_ERRORS = {},
 }
 
 local LONGHAND_FLAGS = {
 	["first-only"] = FLAGS.FIRST_ONLY,
-	["errors"] = FLAGS.DO_PRINT_ERRORS,
-	["text"] = FLAGS.INNER_TEXT,
+	["quiet"] = FLAGS.NO_PRINT_ERRORS
 }
 
 local SHORTHAND_FLAGS = {
-	["1"] = FLAGS.FIRST_ONLY,
-	["e"] = FLAGS.DO_PRINT_ERRORS,
-	["t"] = FLAGS.INNER_TEXT,
+	["f"] = FLAGS.FIRST_ONLY,
+	["q"] = FLAGS.NO_PRINT_ERRORS,
 }
 
 
@@ -98,7 +94,7 @@ for _, argument in ipairs(arg) do
 end
 
 
-if flags[ FLAGS.DO_PRINT_ERRORS ] then
+if not flags[ FLAGS.NO_PRINT_ERRORS ] then
 	logger.enable_printing_errors()
 end
 
@@ -126,7 +122,7 @@ if html_file ~= "-" then
 
 	html = handle:read("a")
 else
-	html = io.read("a")
+	html = io.read()
 end
 
 local document = HTML.parse( html )
@@ -214,12 +210,6 @@ end
 
 if flags[FLAGS.FIRST_ONLY] then
 	if #elements > 0 then
-
-		if flags[FLAGS.INNER_TEXT] then
-			logger.print( elements[1]:inner_text() )
-			return 0
-		end
-
 		logger.print( HTML.tostring( elements[1] ) )
 	end
 
@@ -227,9 +217,5 @@ if flags[FLAGS.FIRST_ONLY] then
 end
 
 for _, el in ipairs(elements) do
-		if flags[FLAGS.INNER_TEXT] then
-			logger.print( el:inner_text() )
-		else
-			logger.print( HTML.tostring(el) )
-		end
+		logger.print( HTML.tostring(el) )
 end