No OneTemporary
Actions

Size

24 KB

Referenced Files

None

Subscribers

None

View Options

	diff --git a/Makefile b/Makefile
	index dab6754..1977293 100644
	--- a/Makefile
	+++ b/Makefile
	@@ -1,65 +1,54 @@
	# SPDX-FileCopyrightText: 2017-2019 myhtmlex authors <https://github.com/Overbryd/myhtmlex>
	# SPDX-FileCopyrightText: 2019-2022 Pleroma Authors <https://pleroma.social>
	# SPDX-License-Identifier: LGPL-2.1-only

	MIX = mix
	CMAKE = cmake
	-CNODE_CFLAGS = -g -O2 -std=c99 -pedantic -Wcomment -Wextra -Wno-old-style-declaration -Wall
	-
	-# ignore unused parameter warnings
	-CNODE_CFLAGS += -Wno-unused-parameter
	+CFLAGS ?= -g -O2 -pedantic -Wcomment -Wextra -Wno-old-style-declaration -Wall

	# set erlang include path
	ERLANG_PATH = $(shell erl -eval 'io:format("~s", [lists:concat([code:root_dir(), "/erts-", erlang:system_info(version)])])' -s init stop -noshell)
	CNODE_CFLAGS += -I$(ERLANG_PATH)/include

	-# expecting myhtml as a submodule in c_src/
	+# expecting lexbor as a submodule in c_src/
	# that way we can pin a version and package the whole thing in hex
	-# hex does not allow for non-app related dependencies.
	LXB_PATH = c_src/lexbor
	-LXB_STATIC = $(LXB_PATH)/liblexbor_static.a
	-CNODE_CFLAGS += -I$(LXB_PATH)/source
	-# avoid undefined reference errors to phtread_mutex_trylock
	-CNODE_CFLAGS += -lpthread
	+LXB_AR = $(LXB_PATH)/liblexbor_static.a
	+ifeq ($(WITH_SYSTEM_LEXBOR),1)
	+ LXB_CFLAGS =
	+ LXB_LDFLAGS = -llexbor
	+ LXB_DEPS =
	+else
	+ LXB_CFLAGS = -I$(LXB_PATH)/source
	+ LXB_LDFLAGS = $(LXB_AR)
	+ LXB_DEPS = $(LXB_AR)
	+endif

	# C-Node
	ERL_INTERFACE = $(wildcard $(ERLANG_PATH)/../lib/erl_interface-*)
	-CNODE_CFLAGS += -L$(ERL_INTERFACE)/lib
	CNODE_CFLAGS += -I$(ERL_INTERFACE)/include
	-
	-CNODE_LDFLAGS =
	-
	-ifeq ($(OTP22_DEF),YES)
	- CNODE_CFLAGS += -DOTP_22_OR_NEWER
	-else
	- CNODE_LDFLAGS += -lerl_interface
	-endif
	-
	-CNODE_LDFLAGS += -lei -pthread
	-
	-CNODE_CFLAGS += $(CPPFLAGS) $(CFLAGS)
	-CNODE_LDFLAGS += $(LDFLAGS)
	+CNODE_LDFLAGS += -L$(ERL_INTERFACE)/lib -lei -lpthread

	.PHONY: all

	all: priv/fasthtml_worker

	-$(LXB_STATIC): $(LXB_PATH)
	+$(LXB_AR): $(LXB_PATH)
	# Sadly, build components separately seems to sporadically fail
	cd $(LXB_PATH); \
	- CFLAGS='$(CPPFLAGS) $(CFLAGS)' \
	+ CFLAGS='$(CFLAGS)' \
	cmake -DLEXBOR_BUILD_SEPARATELY=OFF -DLEXBOR_BUILD_SHARED=OFF
	$(MAKE) -C $(LXB_PATH)

	-priv/fasthtml_worker: c_src/fasthtml_worker.c $(LXB_STATIC)
	+priv/fasthtml_worker: c_src/fasthtml_worker.c $(LXB_DEPS)
	mkdir -p priv
	- $(CC) -o $@ $< $(LXB_STATIC) $(CNODE_CFLAGS) $(CNODE_LDFLAGS)
	+ $(CC) -std=c99 $(CFLAGS) $(CNODE_CFLAGS) $(LXB_CFLAGS) -o $@ $< $(LDFLAGS) $(CNODE_LDFLAGS) $(LXB_LDFLAGS)

	clean: clean-myhtml
	$(RM) -r priv/myhtmlex*
	$(RM) priv/fasthtml_worker
	$(RM) myhtmlex-*.tar
	$(RM) -r package-test

	clean-myhtml:
	- $(MAKE) -C $(MYHTML_PATH) clean
	+ $(MAKE) -C $(LXB_PATH) clean
	diff --git a/README.md b/README.md
	index 391e168..572c826 100644
	--- a/README.md
	+++ b/README.md
	@@ -1,34 +1,42 @@
	<!--
	SPDX-FileCopyrightText: 2017-2019 myhtmlex authors <https://github.com/Overbryd/myhtmlex>
	SPDX-FileCopyrightText: 2019-2022 Pleroma Authors <https://pleroma.social>
	SPDX-License-Identifier: LGPL-2.1-only
	-->

	# FastHTML

	-A C Node wrapping lexborisov's [myhtml](https://github.com/lexborisov/myhtml).
	+A C Node wrapping [lexbor](https://github.com/lexbor/lexbor).
	Primarily used with [FastSanitize](https://git.pleroma.social/pleroma/fast_sanitize).

	* Available as a hex package: `{:fast_html, "~> 2.0"}`
	* [Documentation](https://hexdocs.pm/fast_html/fast_html.html)

	+## Compiling
	+- GNU Make
	+- C Compiler
	+- Erlang 22.0+ with development headers
	+- (optional) [lexbor](https://github.com/lexbor/lexbor) 2.2.0+
	+
	+If you want to use a system installation of lexbor, you can set `WITH_SYSTEM_LEXBOR=1` during compilation time. By default it will used the vendored version present at `c_src/lexbor`.
	+
	## Benchmarks

	The following table provides median times it takes to decode a string to a tree for html parsers that can be used from Elixir. Benchmarks were conducted on a machine with an `AMD Ryzen 9 3950X (32) @ 3.500GHz` CPU and 32GB of RAM. The `mix fast_html.bench` task can be used for running the benchmark by yourself.

	\| File/Parser \| fast_html (Port) \| mochiweb_html (erlang) \| html5ever (Rust NIF) \| Myhtmlex (NIF)¹ \|
	\|----------------------\|--------------------\|------------------------\|----------------------\|----------------\|
	\| document-large.html (6.9M) \| 125.12 ms \| 1778.34 ms \| 395.21 ms \| 327.17 ms \|
	\| document-small.html (25K)\| 0.50 ms \| 2.76 ms \| 1.72 ms \| 1.19 ms \|
	\| fragment-large.html (33K)\| 0.93 ms \| 4.78 ms \| 2.34 ms \| 2.15 ms \|
	\| fragment-small.html² (757B)\| 44.60 μs \| 42.13 μs \| 43.58 μs \| 289.71 μs \|

	Full benchmark output can be seen in [this snippet](https://git.pleroma.social/pleroma/elixir-libraries/fast_html/snippets/3128)

	1. Myhtmlex has a C-Node mode, but it wasn't benchmarked here because it segfaults on `document-large.html`
	2. The slowdown on `fragment-small.html` is due to Port overhead. Unlike html5ever and Myhtmlex in NIF mode, `fast_html` has the parser process isolated and communicates with it over stdio, so even if a fatal crash in the parser happens, it won't bring down the entire VM.

	## Contribution / Bug Reports

	* Please make sure you do `git submodule update` after a checkout/pull
	* The project aims to be fully tested
	diff --git a/c_src/fasthtml_worker.c b/c_src/fasthtml_worker.c
	index ad2baa2..c0921b1 100644
	--- a/c_src/fasthtml_worker.c
	+++ b/c_src/fasthtml_worker.c
	@@ -1,537 +1,530 @@
	// SPDX-FileCopyrightText: 2019-2022 Pleroma Authors <https://pleroma.social>
	// SPDX-License-Identifier: LGPL-2.1-only

	#include <stdlib.h>
	#include <stdbool.h>
	#include <stdio.h>
	#include <string.h>
	#include <stdarg.h>
	#include <unistd.h>
	#include <sys/types.h>
	#include <sys/socket.h>
	#include <netinet/in.h>
	#include <arpa/inet.h>
	#include <errno.h>
	#include <ctype.h>

	#ifndef _REENTRANT
	#define _REENTRANT /* For some reason __erl_errno is undefined unless _REENTRANT is defined */
	#endif

	#include "ei.h"
	-#ifndef OTP_22_OR_NEWER
	-# include "erl_interface.h"
	-#endif

	#define HEADER_SIZE 4

	#include <lexbor/html/html.h>
	#include "tstack.h"

	#ifdef __GNUC__
	# define AFP(x, y) __attribute__((format (printf, x, y)))
	#else
	# define AFP(x, y)
	#endif

	#ifdef __GNUC__
	# define NORETURN __attribute__((noreturn))
	#else
	# define NORETURN
	#endif

	typedef struct _state_t {
	ei_x_buff buffer;
	} state_t;

	typedef enum parse_flags_e {
	FLAG_HTML_ATOMS = 1 << 0,
	FLAG_NIL_SELF_CLOSING = 1 << 1,
	FLAG_COMMENT_TUPLE3 = 1 << 2
	} parse_flags_t;

	char* read_packet(int *len);
	static void handle_send(state_t * state);
	static void err_term(ei_x_buff * response, const char * error_atom);
	static parse_flags_t decode_parse_flags(state_t * state, int arity);
	static void decode(state_t * state, ei_x_buff * response, lxb_html_document_t document, bool fragment, lxb_dom_element_t context_element, lxb_char_t * bin_data, size_t bin_size, parse_flags_t parse_flags);

	static void build_tree(ei_x_buff * response, lxb_dom_node_t* tree, parse_flags_t parse_flags);
	static void prepare_node_attrs(ei_x_buff * response, lxb_dom_node_t* node);

	static inline char * lowercase(char * c);

	static void panic(const char *fmt, ...) AFP(1, 2);
	static void panic(const char *fmt, ...) {
	char buf[4096];
	va_list va;

	va_start (va, fmt);
	vsnprintf (buf, sizeof buf, fmt, va);
	va_end (va);

	fprintf (stderr, "fast_html worker: error: %s\n", buf);
	exit (EXIT_FAILURE);
	}

	int main(int argc, const char *argv[]) {
	state_t* state = calloc (1, sizeof(state_t));

	-#ifdef OTP_22_OR_NEWER
	// initialize erlang client library
	ei_init ();
	-#else
	- erl_init (NULL, -1);
	-#endif

	ei_x_new (&state->buffer);

	fflush (stdout);

	while (true) {
	int len;
	char* buf = read_packet(&len);
	ei_x_free(&state->buffer);
	state->buffer.index = 0;
	state->buffer.buff = buf;
	state->buffer.buffsz = len;
	handle_send (state);
	}

	// shutdown: free all state
	ei_x_free (&state->buffer);
	free (state);

	return EXIT_SUCCESS;
	}


	/*
	* Reads a packet from Erlang. The packet must be a standard {packet, 2}
	* packet. This function aborts if any error is detected (including EOF).
	*
	* Returns: The number of bytes in the packet.
	*/

	char read_packet(int len)
	{

	char* io_buf = NULL; /* Buffer for file i/o. */
	unsigned char header[HEADER_SIZE];
	uint32_t packet_length; /* Length of current packet. */
	uint32_t bytes_read;
	uint32_t total_bytes_read;

	/*
	* Read the packet header.
	*/

	total_bytes_read = read(STDIN_FILENO, header, HEADER_SIZE);

	if (total_bytes_read == 0) {
	exit(0);
	}
	if (total_bytes_read != HEADER_SIZE) {
	panic("Failed to read packet header, read: %d\n", total_bytes_read);
	}

	/*
	* Get the length of this packet.
	*/

	packet_length = 0;

	for (int i = 0; i < HEADER_SIZE; i++)
	packet_length = (packet_length << 8) \| header[i];

	*len=packet_length;

	if ((io_buf = (char *) malloc(packet_length)) == NULL) {
	panic("insufficient memory for i/o buffer of size %d\n", packet_length);
	}

	/*
	* Read the packet itself.
	*/

	total_bytes_read = 0;

	while((bytes_read = read(STDIN_FILENO, (io_buf + total_bytes_read), (packet_length - total_bytes_read))))
	total_bytes_read += bytes_read;

	if (total_bytes_read != packet_length) {
	free(io_buf);
	panic("couldn't read packet of length %d, read: %d\r\n",
	packet_length, total_bytes_read);
	}

	return io_buf;
	}

	// handle ERL_SEND message type.
	// we expect a tuple with arity of 3 or 4 in state->buffer.
	// we expect the first argument to be an atom (`decode` or `decode_fragment`),
	// the second argument to be the HTML payload, and the
	// third argument to be the argument list.
	// In case of `decode_fragment`, the fourth argument should be
	// the context tag name.
	// any other message: respond with an {error, unknown_call} tuple.
	static void handle_send (state_t * state)
	{
	// response holds our response, prepare it
	ei_x_buff response;

	ei_x_new (&response);

	// check the protocol version, if it's unsupported, panic
	int version;
	if (ei_decode_version (state->buffer.buff, &state->buffer.index, &version) < 0)
	panic ("malformed message - bad version (%d).", version);

	// decode the tuple header
	int arity;
	if (ei_decode_tuple_header (state->buffer.buff, &state->buffer.index, &arity) < 0)
	{
	err_term (&response, "badmatch");
	goto out;
	}

	char atom[MAXATOMLEN];
	if (ei_decode_atom (state->buffer.buff, &state->buffer.index, atom) < 0)
	{
	err_term (&response, "badmatch");
	goto out;
	}

	bool fragment = false;
	if (strcmp (atom, "decode"))
	{
	if (strcmp (atom, "decode_fragment")) {
	err_term (&response, "unknown_call");
	goto out;
	} else if (arity != 4) {
	err_term (&response, "badmatch");
	goto out;
	} else {
	fragment = true;
	}
	} else if (arity != 3) {
	err_term (&response, "badmatch");
	goto out;
	}

	// the next argument should be a binary, allocate it dynamically.
	int bin_type, bin_size;
	if (ei_get_type (state->buffer.buff, &state->buffer.index, &bin_type, &bin_size) < 0)
	panic ("failed to decode binary size in message");

	// verify the type
	if (bin_type != ERL_BINARY_EXT)
	{
	err_term (&response, "badmatch");
	goto out;
	}

	// decode the binary
	char * bin_data = calloc (1, bin_size + 1);
	if (ei_decode_binary (state->buffer.buff, &state->buffer.index, bin_data, NULL) < 0)
	panic ("failed to decode binary in message");

	// next should be the options list
	if (ei_decode_list_header (state->buffer.buff, &state->buffer.index, &arity) < 0)
	panic ("failed to decode options list header in message");

	parse_flags_t parse_flags = decode_parse_flags (state, arity);

	// Lists with items always have an empty list as their tail
	if (arity != 0)
	if (ei_decode_list_header (state->buffer.buff, &state->buffer.index, &arity) < 0)
	panic ("failed to decode empty list header after option list in message");

	lxb_html_document_t *document = lxb_html_document_create();
	lxb_dom_element_t *context_element = NULL;

	// if we are parsing a fragment, context tag name should come next
	if (fragment) {
	int context_bin_type, context_bin_size;
	if (ei_get_type (state->buffer.buff, &state->buffer.index, &context_bin_type, &context_bin_size) < 0)
	panic ("failed to decode binary size in message");

	// verify the type
	if (context_bin_type != ERL_BINARY_EXT)
	{
	err_term (&response, "badmatch");
	goto out;
	}

	// decode the binary
	char* context_bin_data = calloc (1, context_bin_size + 1);
	if (ei_decode_binary (state->buffer.buff, &state->buffer.index, context_bin_data, NULL) < 0)
	panic ("failed to decode context binary in message");

	context_element = lxb_dom_document_create_element(&document->dom_document, (lxb_char_t*) context_bin_data, context_bin_size, NULL);
	free (context_bin_data);
	}

	if (context_element && lxb_dom_element_tag_id(context_element) >= LXB_TAG__LAST_ENTRY) {
	err_term (&response, "unknown_context_tag");
	} else {
	decode (state, &response, document, fragment, context_element, (lxb_char_t *) bin_data, bin_size, parse_flags);
	}
	lxb_html_document_destroy(document);
	free (bin_data);

	out: ;
	// send response
	unsigned char header[HEADER_SIZE];
	uint32_t size = (uint32_t) response.index;

	for (int i = HEADER_SIZE-1; i != -1; i--) {
	header[i] = (unsigned char) size & 0xFF;
	size = size >> 8;
	}

	if(write(STDOUT_FILENO, header, sizeof(header)) <0) panic ("Failed to write header to stdout");
	if(write(STDOUT_FILENO, response.buff, response.index) <0) panic ("Failed to write response to stdout");
	// free response
	ei_x_free (&response);
	return;
	}

	static void err_term (ei_x_buff * response, const char * error_atom)
	{
	response->index = 0;
	ei_x_encode_version (response);
	ei_x_encode_tuple_header (response, 2);
	ei_x_encode_atom (response, "error");
	ei_x_encode_atom (response, error_atom);
	}

	static parse_flags_t decode_parse_flags (state_t * state, int arity)
	{
	parse_flags_t parse_flags = 0;

	for (int i = 0; i < arity; i++)
	{
	char atom[MAXATOMLEN];

	if (ei_decode_atom (state->buffer.buff, &state->buffer.index, atom) < 0)
	continue;

	if (! strcmp ("html_atoms", atom))
	parse_flags \|= FLAG_HTML_ATOMS;
	else if (! strcmp ("nil_self_closing", atom))
	parse_flags \|= FLAG_NIL_SELF_CLOSING;
	else if (! strcmp ("comment_tuple3", atom))
	parse_flags \|= FLAG_COMMENT_TUPLE3;
	}

	return parse_flags;
	}

	static void decode(state_t * state, ei_x_buff * response, lxb_html_document_t document, bool fragment, lxb_dom_element_t context_element, lxb_char_t * bin_data, size_t bin_size, parse_flags_t parse_flags)
	{
	// parse tree
	lxb_status_t status;
	lxb_dom_node_t *node;

	if (fragment) {
	node = lxb_html_document_parse_fragment(document, context_element, bin_data, bin_size);
	status = (node == NULL)? LXB_STATUS_ERROR : LXB_STATUS_OK;
	} else {
	status = lxb_html_document_parse(document, bin_data, bin_size);
	node = lxb_dom_interface_node(document);
	}

	if (status != LXB_STATUS_OK)
	{
	err_term (response, "parse_failed");
	return;
	}

	// build tree
	build_tree (response, node, parse_flags);
	}

	// a tag is sent as a tuple:
	// - a string or atom for the tag name
	// - an attribute list
	// - a children list
	// in this function, we prepare the atom and complete attribute list
	static void prepare_tag_header (ei_x_buff * response, const char * tag_string, lxb_dom_node_t* node, parse_flags_t parse_flags)
	{
	lxb_tag_id_t tag_id = lxb_dom_node_tag_id(node);

	ei_x_encode_tuple_header (response, 3);

	if (! (parse_flags & FLAG_HTML_ATOMS) \|\| (tag_id == LXB_TAG__UNDEF \|\| tag_id >= LXB_TAG__LAST_ENTRY))
	ei_x_encode_binary (response, tag_string, strlen (tag_string));
	else
	ei_x_encode_atom (response, tag_string);

	prepare_node_attrs (response, node);
	}

	// prepare an attribute node
	static void prepare_node_attrs(ei_x_buff * response, lxb_dom_node_t* node)
	{
	lxb_dom_attr_t *attr;

	for (attr = lxb_dom_element_first_attribute(lxb_dom_interface_element(node)); attr != NULL; attr = lxb_dom_element_next_attribute(attr))
	{
	size_t attr_name_len;
	char attr_name = (char) lxb_dom_attr_qualified_name(attr, &attr_name_len);
	size_t attr_value_len;
	const char attr_value = (char) lxb_dom_attr_value(attr, &attr_value_len);

	/* guard against poisoned attribute nodes */
	if (! attr_name_len)
	continue;

	ei_x_encode_list_header (response, 1);
	ei_x_encode_tuple_header (response, 2);
	ei_x_encode_binary (response, attr_name, attr_name_len);

	ei_x_encode_binary (response, attr_value, attr_value_len);
	}

	ei_x_encode_empty_list (response);
	}

	// dump a comment node
	static void prepare_comment (ei_x_buff * response, const char * node_comment, size_t comment_len, parse_flags_t parse_flags)
	{
	ei_x_encode_tuple_header (response, parse_flags & FLAG_COMMENT_TUPLE3 ? 3 : 2);
	ei_x_encode_atom (response, "comment");

	if (parse_flags & FLAG_COMMENT_TUPLE3)
	ei_x_encode_list_header (response, 0);

	ei_x_encode_binary (response, node_comment, comment_len);
	}

	#ifdef DEBUG_LIST_MANIP

	#define EMIT_LIST_HDR \
	printf ("list hdr for node %p\n", current_node); \
	fflush (stdout); \
	ei_x_encode_list_header (response, 1)

	#define EMIT_EMPTY_LIST_HDR \
	printf ("list empty for node %p\n", current_node); \
	fflush (stdout); \
	ei_x_encode_list_header (response, 0)

	#define EMIT_LIST_TAIL \
	printf ("list tail for node %p\n", current_node); \
	fflush (stdout); \
	ei_x_encode_empty_list (response)

	#else

	#define EMIT_LIST_HDR ei_x_encode_list_header (response, 1)
	#define EMIT_EMPTY_LIST_HDR ei_x_encode_list_header (response, 0)
	#define EMIT_LIST_TAIL ei_x_encode_empty_list (response)

	#endif

	static void build_tree (ei_x_buff * response, lxb_dom_node_t* node, parse_flags_t parse_flags)
	{
	tstack stack;
	tstack_init (&stack, 30);
	tstack_push (&stack, node);

	lxb_dom_node_t* current_node = node->first_child;

	// ok we're going to send an actual response so start encoding it
	response->index = 0;
	ei_x_encode_version (response);
	ei_x_encode_tuple_header(response, 2);
	ei_x_encode_atom(response, "ok");

	if (current_node == NULL) {
	EMIT_EMPTY_LIST_HDR;
	EMIT_LIST_TAIL;
	}
	while (current_node != NULL)
	{
	if (current_node->type == LXB_DOM_NODE_TYPE_TEXT)
	{
	size_t text_len;
	const char * node_text = (char*) lxb_dom_node_text_content(current_node, &text_len);
	EMIT_LIST_HDR;
	ei_x_encode_binary (response, node_text, text_len);
	}
	else if (current_node->type == LXB_DOM_NODE_TYPE_COMMENT)
	{
	size_t comment_len;
	const char* node_comment = (char*) lxb_dom_node_text_content(current_node, &comment_len);

	EMIT_LIST_HDR;
	prepare_comment(response, node_comment, comment_len, parse_flags);
	}
	else if(current_node->type == LXB_DOM_NODE_TYPE_ELEMENT)
	{
	// get name of tag
	size_t tag_name_len;
	const char tag_name = (char) lxb_dom_element_qualified_name(lxb_dom_interface_element(current_node), &tag_name_len);
	EMIT_LIST_HDR;
	prepare_tag_header (response, tag_name, current_node, parse_flags);

	if (current_node->first_child)
	{
	tstack_push (&stack, current_node);
	current_node = current_node->first_child;

	continue;
	}
	else
	{
	if (parse_flags & FLAG_NIL_SELF_CLOSING && lxb_html_tag_is_void(lxb_dom_node_tag_id(current_node))) {

	#ifdef DEBUG_LIST_MANIP
	printf ("self-closing tag %s emit nil?\n", tag_string); fflush (stdout);
	#endif
	ei_x_encode_atom (response, "nil");
	}
	else
	{
	EMIT_EMPTY_LIST_HDR;
	}
	}
	}

	if (current_node->next)
	current_node = current_node->next;
	else
	{
	while (! current_node->next && stack.used != 0)
	{
	EMIT_LIST_TAIL;
	current_node = tstack_pop (&stack);
	}

	if (current_node->next)
	current_node = current_node->next;
	}

	// are we at root?
	if (current_node == node)
	break;
	}

	tstack_free (&stack);
	}

	static inline char * lowercase(char* c)
	{
	char * p = c;

	while (*p)
	{
	p = tolower ((unsigned char) p);
	p++;
	}

	return c;
	}
	diff --git a/mix.exs b/mix.exs
	index 9698198..4f042a7 100644
	--- a/mix.exs
	+++ b/mix.exs
	@@ -1,117 +1,98 @@
	# SPDX-FileCopyrightText: 2017-2019 myhtmlex authors <https://github.com/Overbryd/myhtmlex>
	# SPDX-FileCopyrightText: 2019-2022 Pleroma Authors <https://pleroma.social>
	# SPDX-License-Identifier: LGPL-2.1-only

	defmodule FastHtml.Mixfile do
	use Mix.Project

	def project do
	[
	app: :fast_html,
	version: "2.2.0",
	elixir: "~> 1.11",
	deps: deps(),
	package: package(),
	compilers: [:elixir_make] ++ Mix.compilers(),
	make_env: make_env(),
	make_error_message: make_error_message(),
	build_embedded: Mix.env() == :prod,
	start_permanent: Mix.env() == :prod,
	name: "FastHtml",
	description: """
	A module to decode HTML into a tree,
	porting all properties of the underlying
	library lexbor, being fast and correct
	in regards to the html spec.
	""",
	docs: docs()
	]
	end

	def package do
	[
	maintainers: ["Ariadne Conill", "rinpatch"],
	licenses: ["LGPL-2.1-only"],
	links: %{
	"GitLab" => "https://git.pleroma.social/pleroma/elixir-libraries/fast_html/",
	"Issues" => "https://git.pleroma.social/pleroma/elixir-libraries/fast_html/issues",
	"lexbor" => "https://github.com/lexbor/lexbor"
	},
	files: hex_files()
	]
	end

	def application do
	[
	extra_applications: [:logger],
	mod: {FastHtml.Application, []}
	]
	end

	defp deps do
	[
	# documentation helpers
	{:ex_doc, "~> 0.19", only: :dev},
	# benchmarking helpers
	{:benchee, "~> 1.0", only: :bench, optional: true},
	{:dialyxir, "~> 1.0", only: [:dev, :test], runtime: false},
	{:myhtmlex, "~> 0.2.0", only: :bench, runtime: false, optional: true},
	{:mochiweb, "~> 2.18", only: :bench, optional: true},
	{:html5ever,
	git: "https://github.com/rusterlium/html5ever_elixir.git", only: :bench, optional: true},
	{:nimble_pool, "~> 0.2.0"},
	{:elixir_make, "~> 0.4", runtime: false}
	]
	end

	defp docs do
	[
	main: "readme",
	extras: ["README.md", "CHANGELOG.md"]
	]
	end

	defp hex_files do
	# This is run every time mix is executed, so it will fail in the hex package,
	# therefore check if git is even available
	if File.exists?(".git") and System.find_executable("git") do
	{files, 0} = System.cmd("git", ["ls-files", "--recurse-submodules"])

	files
	\|> String.split("\n")
	# Last element is "", which makes hex include all files in the folder to the project
	\|> List.delete_at(-1)
	\|> Enum.reject(fn path ->
	Path.dirname(path) == "bench_fixtures" or
	(Path.dirname(path) != "priv" and String.starts_with?(Path.basename(path), "."))
	end)
	else
	[]
	end
	end

	- defp otp_version do
	- :erlang.system_info(:otp_release)
	- \|> to_string()
	- \|> String.to_integer()
	- end
	-
	- defp otp_22_or_newer? do
	- otp_version() >= 22
	- end
	-
	- defp make_env do
	- %{
	- "OTP22_DEF" =>
	- if otp_22_or_newer?() do
	- "YES"
	- else
	- "NO"
	- end
	- }
	- end
	+ defp make_env, do: %{}

	defp make_error_message,
	do:
	- "Please check you have: a C compiler, GNU\Make, CMake and Erlang development headers installed before reporting an issue."
	+ "Please check you have: a C compiler, GNU Make, CMake and Erlang development headers installed before reporting an issue."
	end

File Metadata

Mime Type: text/x-diff
Expires: Sun, Nov 24, 1:45 PM (1 d, 3 h)
Storage Engine: blob
Storage Format: Raw Data
Storage Handle: 39015
Default Alt Text: (24 KB)

No OneTemporaryActions

View Options

File Metadata

Event Timeline

No OneTemporary
Actions