commit c566966: [Minor] Remove ragel based received parser
Vsevolod Stakhov
vsevolod at highsecure.ru
Thu Feb 7 15:14:20 UTC 2019
Author: Vsevolod Stakhov
Date: 2019-02-07 15:13:35 +0000
URL: https://github.com/rspamd/rspamd/commit/c56696612bb1e20fe907f6285866c4cb841a38e4 (HEAD -> master)
[Minor] Remove ragel based received parser
---
src/CMakeLists.txt | 1 -
src/ragel/smtp_received.rl | 61 -------
src/ragel/smtp_received_parser.rl | 327 --------------------------------------
3 files changed, 389 deletions(-)
diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
index 71ce71119..6fbfa577d 100644
--- a/src/CMakeLists.txt
+++ b/src/CMakeLists.txt
@@ -118,7 +118,6 @@ SET(RAGEL_DEPENDS "${CMAKE_SOURCE_DIR}/src/ragel/smtp_address.rl"
"${CMAKE_SOURCE_DIR}/src/ragel/smtp_date.rl"
"${CMAKE_SOURCE_DIR}/src/ragel/smtp_ip.rl"
"${CMAKE_SOURCE_DIR}/src/ragel/smtp_base.rl"
- "${CMAKE_SOURCE_DIR}/src/ragel/smtp_received.rl"
"${CMAKE_SOURCE_DIR}/src/ragel/content_disposition.rl")
RAGEL_TARGET(ragel_smtp_addr
INPUTS ${CMAKE_SOURCE_DIR}/src/ragel/smtp_addr_parser.rl
diff --git a/src/ragel/smtp_received.rl b/src/ragel/smtp_received.rl
deleted file mode 100644
index 7635fcee4..000000000
--- a/src/ragel/smtp_received.rl
+++ /dev/null
@@ -1,61 +0,0 @@
-%%{
- machine smtp_received;
-
-
- # http://tools.ietf.org/html/rfc5321#section-4.4
-
- Addtl_Link = Atom;
- Link = "TCP" | Addtl_Link;
- Attdl_Protocol = Atom;
- Protocol = "ESMTP"i %ESMTP_proto |
- "SMTP"i %SMTP_proto |
- "ESMTPS"i %ESMTPS_proto |
- "ESMTPA"i %ESMTPA_proto |
- "ESMTPSA"i %ESMTPSA_proto |
- "LMTP"i %LMTP_proto |
- "IMAP"i %IMAP_proto |
- Attdl_Protocol;
-
- TCP_info = address_literal >Real_IP_Start %Real_IP_End |
- ( Domain >Real_Domain_Start %Real_Domain_End FWS address_literal >Real_IP_Start %Real_IP_End ) |
- ( non_conformant_address_literal >Real_IP_Start %Real_IP_End );
- Extended_Domain = (Domain >Real_Domain_Start %Real_Domain_End | # Used to be a real domain
- ( Domain >Reported_Domain_Start %Reported_Domain_End FWS "(" TCP_info ")" ) | # Here domain is something specified by remote side
- ( address_literal >Real_Domain_Start %Real_Domain_End FWS "(" TCP_info ")" ) );
-
- ccontent = ctext | FWS | '(' @{ fcall balanced_ccontent; };
- balanced_ccontent := ccontent* ')' @{ fret; };
- comment = "(" ((WSP* ccontent)* WSP*) >Comment_Start %Comment_End ")";
- CFWS = WSP* (comment+ WSP*)*;
-
- From_domain = "FROM"i FWS Extended_Domain >From_Start %From_End;
- By_domain = "BY"i FWS Extended_Domain >By_Start %By_End;
-
- Retarded_Domain = TCP_info;
- From_domain_retarded = "FROM"i FWS Retarded_Domain >From_Start %From_End;
-
- Via = CFWS "VIA"i FWS Link;
- With = CFWS "WITH"i FWS Protocol;
-
- id_left = dot_atom_text;
- no_fold_literal = "[" dtext* "]";
- id_right = dot_atom_text | no_fold_literal;
- msg_id = "<" id_left "@" id_right ">";
- ID = CFWS "ID"i FWS ( Dot_string | msg_id );
-
- For = CFWS "FOR"i FWS ( Path | Mailbox ) >For_Start %For_End;
- Additional_Registered_Clauses = CFWS Atom FWS String;
- Opt_info = Via? With? ID? For? Additional_Registered_Clauses?;
- # Here we make From part optional just because many received headers lack it
- Received = From_domain? CFWS? By_domain? CFWS? Opt_info CFWS? ";" FWS date_time >Date_Start %Date_End CFWS?;
- Received_retarded = From_domain_retarded CFWS? By_domain? CFWS? Opt_info CFWS? ";" FWS date_time >Date_Start %Date_End CFWS?;
-
- prepush {
- if (top >= st_storage.size) {
- st_storage.size = (top + 1) * 2;
- st_storage.data = realloc (st_storage.data, st_storage.size * sizeof (int));
- g_assert (st_storage.data != NULL);
- stack = st_storage.data;
- }
- }
-}%%
diff --git a/src/ragel/smtp_received_parser.rl b/src/ragel/smtp_received_parser.rl
deleted file mode 100644
index 7c747f9f5..000000000
--- a/src/ragel/smtp_received_parser.rl
+++ /dev/null
@@ -1,327 +0,0 @@
-%%{
-
- machine smtp_received_parser;
-
-
- action IP6_start {
- in_v6 = 1;
- ip_start = p;
- }
- action IP6_end {
- in_v6 = 0;
- ip_end = p;
- }
- action IP4_start {
- if (!in_v6) {
- ip_start = p;
- }
- }
- action IP4_end {
- if (!in_v6) {
- ip_end = p;
- }
- }
-
- action User_start {
- addr->user = p;
- }
-
- action User_end {
- if (addr->user) {
- addr->user_len = p - addr->user;
- }
- }
-
- action Domain_start {
- addr->domain = p;
- }
-
- action Domain_end {
- if (addr->domain) {
- addr->domain_len = p - addr->domain;
- }
- }
-
- action Domain_addr_start {
- addr->domain = p;
- addr->flags |= RSPAMD_EMAIL_ADDR_IP;
- }
-
- action Domain_addr_end {
- if (addr->domain) {
- addr->domain_len = p - addr->domain;
- }
- }
-
- action User_has_backslash {
- addr->flags |= RSPAMD_EMAIL_ADDR_HAS_BACKSLASH;
- }
-
- action Quoted_addr {
- addr->flags |= RSPAMD_EMAIL_ADDR_QUOTED;
- }
-
- action Empty_addr {
- addr->flags |= RSPAMD_EMAIL_ADDR_EMPTY;
- addr->addr = "";
- addr->user = addr->addr;
- addr->domain = addr->addr;
- }
-
- action Valid_addr {
- addr->flags |= RSPAMD_EMAIL_ADDR_VALID;
- }
-
- action Addr_has_angle {
- addr->flags |= RSPAMD_EMAIL_ADDR_BRACED;
- }
-
- action Addr_start {
- addr->addr = p;
- }
-
- action Addr_end {
- if (addr->addr) {
- addr->addr_len = p - addr->addr;
- }
- }
-
- action Real_Domain_Start {
- real_domain_start = p;
- }
- action Real_Domain_End {
- real_domain_end = p;
- }
- action Reported_Domain_Start {
- reported_domain_start = p;
- }
- action Reported_Domain_End {
- reported_domain_end = p;
- }
-
- action Real_IP_Start {
- if (real_ip_end == NULL && real_ip_start == NULL) {
- real_ip_start = p;
- }
- }
- action Real_IP_End {
- if (real_ip_end == NULL && real_ip_start != NULL) {
- if (ip_start && ip_end && ip_end > ip_start) {
- real_ip_start = ip_start;
- real_ip_end = ip_end;
- }
- else {
- real_ip_end = p;
- }
- }
-
- ip_start = NULL;
- ip_end = NULL;
- }
-
- action From_Start {
- real_domain_start = NULL;
- real_domain_end = NULL;
- reported_domain_start = NULL;
- reported_domain_end = NULL;
- ip_start = NULL;
- ip_end = NULL;
- for_start = NULL;
- for_end = NULL;
- }
-
- action By_Start {
- real_domain_start = NULL;
- real_domain_end = NULL;
- reported_domain_start = NULL;
- reported_domain_end = NULL;
- ip_start = NULL;
- ip_end = NULL;
- for_start = NULL;
- for_end = NULL;
- }
-
- action By_End {
- if (real_domain_end && real_domain_start && real_domain_end > real_domain_start) {
- tmplen = real_domain_end - real_domain_start;
- rh->by_hostname = rspamd_mempool_alloc (task->task_pool, tmplen + 1);
- rspamd_strlcpy (rh->by_hostname, real_domain_start, tmplen + 1);
- }
- else if (reported_domain_end && reported_domain_start && reported_domain_end > reported_domain_start) {
- len = reported_domain_end - reported_domain_start;
- rh->by_hostname = rspamd_mempool_alloc (task->task_pool, tmplen + 1);
- rspamd_strlcpy (rh->by_hostname, reported_domain_start, tmplen + 1);
- }
- }
-
- action From_End {
- if (real_domain_end && real_domain_start && real_domain_end > real_domain_start) {
- tmplen = real_domain_end - real_domain_start;
- rh->real_hostname = rspamd_mempool_alloc (task->task_pool, tmplen + 1);
- rspamd_strlcpy (rh->real_hostname, real_domain_start, tmplen + 1);
- }
- if (reported_domain_end && reported_domain_start && reported_domain_end > reported_domain_start) {
- tmplen = reported_domain_end - reported_domain_start;
- rh->from_hostname = rspamd_mempool_alloc (task->task_pool, tmplen + 1);
- rspamd_strlcpy (rh->from_hostname, reported_domain_start, tmplen + 1);
- }
- }
-
- action For_Start {
- for_start = p;
- }
-
- action For_End {
- if (for_start && p > for_start) {
- for_end = p;
- tmplen = for_end - for_start;
- rh->for_mbox = rspamd_mempool_alloc (task->task_pool, tmplen + 1);
- rspamd_strlcpy (rh->for_mbox, for_start, tmplen + 1);
- }
- }
-
- action SMTP_proto {
- rh->type = RSPAMD_RECEIVED_SMTP;
- }
- action ESMTPS_proto {
- rh->type = RSPAMD_RECEIVED_ESMTPS;
- }
- action ESMTPA_proto {
- rh->type = RSPAMD_RECEIVED_ESMTPA;
- }
- action ESMTP_proto {
- rh->type = RSPAMD_RECEIVED_ESMTP;
- }
- action ESMTPSA_proto {
- rh->type = RSPAMD_RECEIVED_ESMTPSA;
- }
- action LMTP_proto {
- rh->type = RSPAMD_RECEIVED_LMTP;
- }
- action IMAP_proto {
- rh->type = RSPAMD_RECEIVED_IMAP;
- }
-
- action Date_Start {
- date_start = p;
- }
- action Date_End {
- if (date_start && p > date_start) {
- rh->timestamp = rspamd_tm_to_time (&tm, tz);
- }
- }
-
- action Comment_Start {
- cstart = p;
- }
-
- action Comment_End {
- cend = p;
-
- if (cend && cstart && cend > cstart) {
- rspamd_smtp_maybe_process_smtp_comment (task, cstart, cend - cstart, rh);
- }
-
- cend = NULL;
- cstart = NULL;
- }
-
- include smtp_base "smtp_base.rl";
- include smtp_ip "smtp_ip.rl";
- include smtp_date "smtp_date.rl";
- include smtp_address"smtp_address.rl";
- include smtp_received "smtp_received.rl";
-
- main := Received;
- retarded := Received_retarded;
-
-}%%
-
-#include "smtp_parsers.h"
-
-%% write data;
-
-int
-rspamd_smtp_received_parse (struct rspamd_task *task, const char *data, size_t len, struct received_header *rh)
-{
- struct rspamd_email_address for_addr, *addr;
- const char *real_domain_start, *real_domain_end,
- *real_ip_start, *real_ip_end,
- *reported_domain_start, *reported_domain_end,
- *ip_start, *ip_end, *date_start,
- *for_start, *for_end, *tmp, *cstart, *cend;
- struct tm tm;
- const char *p = data, *pe = data + len, *eof;
- int cs, in_v6 = 0, *stack = NULL;
- gsize top = 0;
- glong tz = 0;
- struct _ragel_st_storage {
- int *data;
- gsize size;
- } st_storage;
- guint tmplen;
- gboolean retarded_checked = FALSE;
-
- memset (&st_storage, 0, sizeof (st_storage));
- memset (rh, 0, sizeof (*rh));
- memset (&tm, 0, sizeof (tm));
- real_domain_start = NULL;
- real_domain_end = NULL;
- real_ip_start = NULL;
- real_ip_end = NULL;
- reported_domain_start = NULL;
- reported_domain_end = NULL;
- ip_start = NULL;
- ip_end = NULL;
- date_start = NULL;
- for_start = NULL;
- for_end = NULL;
- cstart = NULL;
- cend = NULL;
- rh->type = RSPAMD_RECEIVED_UNKNOWN;
-
- memset (&for_addr, 0, sizeof (for_addr));
- addr = &for_addr;
- eof = pe;
-
- %% write init;
-reexec_retarded:
- %% write exec;
- %% write exports;
-
- if (!real_ip_end && !retarded_checked) {
- cs = smtp_received_parser_en_retarded;
- retarded_checked = TRUE;
- goto reexec_retarded;
- }
-
- if (real_ip_end && real_ip_start && real_ip_end > real_ip_start) {
- tmplen = real_ip_end - real_ip_start;
- rh->real_ip = rspamd_mempool_alloc (task->task_pool, tmplen + 1);
- rspamd_strlcpy (rh->real_ip, real_ip_start, tmplen + 1);
- }
-
- if (!rh->real_ip && rh->comment_ip) {
- rh->real_ip = rh->comment_ip;
- }
-
- if (rh->real_ip && !rh->from_ip) {
- rh->from_ip = rh->real_ip;
- }
- if (rh->real_hostname && !rh->from_hostname) {
- rh->from_hostname = rh->real_hostname;
- }
-
- if (rh->real_ip) {
- if (rspamd_parse_inet_address (&rh->addr, rh->real_ip, strlen (rh->real_ip))) {
- rspamd_mempool_add_destructor (task->task_pool,
- (rspamd_mempool_destruct_t)rspamd_inet_address_free, rh->addr);
- }
- }
-
- if (st_storage.data) {
- free (st_storage.data);
- }
-
- return cs;
-}
More information about the Commits
mailing list