Add a replacement for the nationalchars module.

7 files changed, 391 insertions, 0 deletions

diff --git a/docs/conf/codepages/ascii.conf.example b/docs/conf/codepages/ascii.conf.example
new file mode 100644
index 000000000..7c5ecdd85
--- /dev/null
+++ b/docs/conf/codepages/ascii.conf.example
@@ -0,0 +1,37 @@
+# This file contains ASCII codepage rules for use with the codepage module.
+
+<codepage name="ascii">
+
+<cpchars index="45">                        # -
+<cpchars begin="48"  end="57">              # 01234567899
+<cpchars begin="65"  end="90"  front="yes"> # ABCDEFGHIJKLMNOPQRSTUVWXYZ
+<cpchars begin="91"  end="96"  front="yes"> # [\]^_`
+<cpchars begin="97"  end="122" front="yes"> # abcdefghijklmnopqrstuvwxyz
+<cpchars begin="123" end="125" front="yes"> # {|}
+
+<cpcase lower="97"  upper="65"> # a => A
+<cpcase lower="98"  upper="66"> # b => B
+<cpcase lower="99"  upper="67"> # c => C
+<cpcase lower="100" upper="68"> # d => D
+<cpcase lower="101" upper="69"> # e => E
+<cpcase lower="102" upper="70"> # f => F
+<cpcase lower="103" upper="71"> # g => G
+<cpcase lower="104" upper="72"> # h => H
+<cpcase lower="105" upper="73"> # i => I
+<cpcase lower="106" upper="74"> # j => J
+<cpcase lower="107" upper="75"> # k => K
+<cpcase lower="108" upper="76"> # l => L
+<cpcase lower="109" upper="77"> # m => M
+<cpcase lower="110" upper="78"> # n => N
+<cpcase lower="111" upper="79"> # o => O
+<cpcase lower="112" upper="80"> # p => P
+<cpcase lower="113" upper="81"> # q => Q
+<cpcase lower="114" upper="82"> # r => R
+<cpcase lower="115" upper="83"> # s => S
+<cpcase lower="116" upper="84"> # t => T
+<cpcase lower="117" upper="85"> # u => U
+<cpcase lower="118" upper="86"> # v => V
+<cpcase lower="119" upper="87"> # w => W
+<cpcase lower="120" upper="88"> # x => X
+<cpcase lower="121" upper="89"> # y => Y
+<cpcase lower="122" upper="90"> # z => Z
diff --git a/docs/conf/codepages/latin1.conf.example b/docs/conf/codepages/latin1.conf.example
new file mode 100644
index 000000000..3beb002fd
--- /dev/null
+++ b/docs/conf/codepages/latin1.conf.example
@@ -0,0 +1,42 @@
+# This file contains ISO 8859-1 codepage rules for use with the codepage module.
+
+<codepage name="latin1">
+
+<cpchars begin="192" end="214" front="yes"> # ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ
+<cpchars begin="216" end="246" front="yes"> # ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö
+<cpchars begin="248" end="255" front="yes"> # øùúûüýþÿ
+
+<cpcase lower="83"  upper="223"> # ß => S
+<cpcase lower="192" upper="224"> # à => À
+<cpcase lower="193" upper="225"> # á => Á
+<cpcase lower="194" upper="226"> # â => Â
+<cpcase lower="195" upper="227"> # ã => Ã
+<cpcase lower="196" upper="228"> # ä => Ä
+<cpcase lower="197" upper="229"> # å => Å
+<cpcase lower="198" upper="230"> # æ => Æ
+<cpcase lower="199" upper="231"> # ç => Ç
+<cpcase lower="200" upper="232"> # è => È
+<cpcase lower="201" upper="233"> # é => É
+<cpcase lower="202" upper="234"> # ê => Ê
+<cpcase lower="203" upper="235"> # ë => Ë
+<cpcase lower="204" upper="236"> # ì => Ì
+<cpcase lower="205" upper="237"> # í => Í
+<cpcase lower="206" upper="238"> # î => Î
+<cpcase lower="207" upper="239"> # ï => Ï
+<cpcase lower="208" upper="240"> # ð => Ð
+<cpcase lower="209" upper="241"> # ñ => Ñ
+<cpcase lower="210" upper="242"> # ò => Ò
+<cpcase lower="211" upper="243"> # ó => Ó
+<cpcase lower="212" upper="244"> # ô => Ô
+<cpcase lower="213" upper="245"> # õ => Õ
+<cpcase lower="214" upper="246"> # ö => Ö
+<cpcase lower="216" upper="248"> # ø => Ø
+<cpcase lower="217" upper="249"> # ù => Ù
+<cpcase lower="218" upper="250"> # ú => Ú
+<cpcase lower="219" upper="251"> # û => Û
+<cpcase lower="220" upper="252"> # ü => Ü
+<cpcase lower="221" upper="253"> # ý => Ý
+<cpcase lower="222" upper="254"> # þ => Þ
+
+# Include the ASCII rules to avoid duplication.
+<include file="examples/codepages/ascii.conf.example">
diff --git a/docs/conf/codepages/rfc1459.conf.example b/docs/conf/codepages/rfc1459.conf.example
new file mode 100644
index 000000000..32f453044
--- /dev/null
+++ b/docs/conf/codepages/rfc1459.conf.example
@@ -0,0 +1,41 @@
+# This file contains RFC 1459 codepage rules for use with the codepage module.
+
+<codepage name="rfc1459">
+
+<cpchars index="45">                        # -
+<cpchars begin="48"  end="57">              # 01234567899
+<cpchars begin="65"  end="90"  front="yes"> # ABCDEFGHIJKLMNOPQRSTUVWXYZ
+<cpchars begin="91"  end="96"  front="yes"> # [\]^_`
+<cpchars begin="97"  end="122" front="yes"> # abcdefghijklmnopqrstuvwxyz
+<cpchars begin="123" end="125" front="yes"> # {|}
+
+<cpcase lower="97"  upper="65"> # a => A
+<cpcase lower="98"  upper="66"> # b => B
+<cpcase lower="99"  upper="67"> # c => C
+<cpcase lower="100" upper="68"> # d => D
+<cpcase lower="101" upper="69"> # e => E
+<cpcase lower="102" upper="70"> # f => F
+<cpcase lower="103" upper="71"> # g => G
+<cpcase lower="104" upper="72"> # h => H
+<cpcase lower="105" upper="73"> # i => I
+<cpcase lower="106" upper="74"> # j => J
+<cpcase lower="107" upper="75"> # k => K
+<cpcase lower="108" upper="76"> # l => L
+<cpcase lower="109" upper="77"> # m => M
+<cpcase lower="110" upper="78"> # n => N
+<cpcase lower="111" upper="79"> # o => O
+<cpcase lower="112" upper="80"> # p => P
+<cpcase lower="113" upper="81"> # q => Q
+<cpcase lower="114" upper="82"> # r => R
+<cpcase lower="115" upper="83"> # s => S
+<cpcase lower="116" upper="84"> # t => T
+<cpcase lower="117" upper="85"> # u => U
+<cpcase lower="118" upper="86"> # v => V
+<cpcase lower="119" upper="87"> # w => W
+<cpcase lower="120" upper="88"> # x => X
+<cpcase lower="121" upper="89"> # y => Y
+<cpcase lower="122" upper="90"> # z => Z
+<cpcase lower="123" upper="91"> # { => [
+<cpcase lower="124" upper="92"> # | => \
+<cpcase lower="125" upper="93"> # } => ]
+<cpcase lower="126" upper="94"> # ~ => ^
diff --git a/docs/conf/codepages/strict-rfc1459.conf.example b/docs/conf/codepages/strict-rfc1459.conf.example
new file mode 100644
index 000000000..b14e477e8
--- /dev/null
+++ b/docs/conf/codepages/strict-rfc1459.conf.example
@@ -0,0 +1,40 @@
+# This file contains strict RFC 1459 codepage rules for use with the codepage module.
+
+<codepage name="rfc1459">
+
+<cpchars index="45">                        # -
+<cpchars begin="48"  end="57">              # 01234567899
+<cpchars begin="65"  end="90"  front="yes"> # ABCDEFGHIJKLMNOPQRSTUVWXYZ
+<cpchars begin="91"  end="96"  front="yes"> # [\]^_`
+<cpchars begin="97"  end="122" front="yes"> # abcdefghijklmnopqrstuvwxyz
+<cpchars begin="123" end="125" front="yes"> # {|}
+
+<cpcase lower="97"  upper="65"> # a => A
+<cpcase lower="98"  upper="66"> # b => B
+<cpcase lower="99"  upper="67"> # c => C
+<cpcase lower="100" upper="68"> # d => D
+<cpcase lower="101" upper="69"> # e => E
+<cpcase lower="102" upper="70"> # f => F
+<cpcase lower="103" upper="71"> # g => G
+<cpcase lower="104" upper="72"> # h => H
+<cpcase lower="105" upper="73"> # i => I
+<cpcase lower="106" upper="74"> # j => J
+<cpcase lower="107" upper="75"> # k => K
+<cpcase lower="108" upper="76"> # l => L
+<cpcase lower="109" upper="77"> # m => M
+<cpcase lower="110" upper="78"> # n => N
+<cpcase lower="111" upper="79"> # o => O
+<cpcase lower="112" upper="80"> # p => P
+<cpcase lower="113" upper="81"> # q => Q
+<cpcase lower="114" upper="82"> # r => R
+<cpcase lower="115" upper="83"> # s => S
+<cpcase lower="116" upper="84"> # t => T
+<cpcase lower="117" upper="85"> # u => U
+<cpcase lower="118" upper="86"> # v => V
+<cpcase lower="119" upper="87"> # w => W
+<cpcase lower="120" upper="88"> # x => X
+<cpcase lower="121" upper="89"> # y => Y
+<cpcase lower="122" upper="90"> # z => Z
+<cpcase lower="123" upper="91"> # { => [
+<cpcase lower="124" upper="92"> # | => \
+<cpcase lower="125" upper="93"> # } => ]
diff --git a/docs/conf/modules.conf.example b/docs/conf/modules.conf.example
index 215f66137..29adff840 100644
--- a/docs/conf/modules.conf.example
+++ b/docs/conf/modules.conf.example
@@ -572,6 +572,20 @@
 #<module name="clones">
 
 #-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#
+# Codepage module: Allows using a custom 8-bit codepage for nicknames
+# and case mapping.
+#
+# You should include one of the following files to set your codepage:
+#<include file="examples/codepages/ascii.conf.example">
+#<include file="examples/codepages/latin1.conf.example">
+#<include file="examples/codepages/rfc1459.conf.example">
+#<include file="examples/codepages/strict-rfc1459.conf.example">
+#
+# You can also define a custom codepage. For details on how to do this
+# please refer to the docs site:
+# https://docs.inspircd.org/3/modules/codepage
+
+#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#-#
 # Common channels module: Adds user mode +c, which, when set, requires
 # that users must share a common channel with you to PRIVMSG or NOTICE
 # you.
@@ -1414,6 +1428,9 @@
 # National characters module:
 # 1) Allows using national characters in nicknames.
 # 2) Allows using custom (national) casemapping over the network.
+#
+# This module is incredibly poorly written and documented. You should
+# probably use the codepage module instead for 8-bit codepages.
 #<module name="nationalchars">
 #
 # file - Location of the file which contains casemapping rules. If this
diff --git a/make/template/main.mk b/make/template/main.mk
index 1e6ba5cd0..f71852aa5 100644
--- a/make/template/main.mk
+++ b/make/template/main.mk
@@ -210,6 +210,7 @@ install: target
 	@-$(INSTALL) -d -g @GID@ -o @UID@ -m $(INSTMODE_DIR) $(BINPATH)
 	@-$(INSTALL) -d -g @GID@ -o @UID@ -m $(INSTMODE_DIR) $(CONPATH)
 	@-$(INSTALL) -d -g @GID@ -o @UID@ -m $(INSTMODE_DIR) $(DATPATH)
+	@-$(INSTALL) -d -g @GID@ -o @UID@ -m $(INSTMODE_DIR) $(EXAPATH)/codepages
 	@-$(INSTALL) -d -g @GID@ -o @UID@ -m $(INSTMODE_DIR) $(EXAPATH)/providers
 	@-$(INSTALL) -d -g @GID@ -o @UID@ -m $(INSTMODE_DIR) $(EXAPATH)/services
 	@-$(INSTALL) -d -g @GID@ -o @UID@ -m $(INSTMODE_DIR) $(EXAPATH)/sql
@@ -230,6 +231,7 @@ endif
 	-$(INSTALL) -g @GID@ -o @UID@ -m $(INSTMODE_TXT) @CONFIGURE_DIRECTORY@/inspircd-genssl.1 $(MANPATH) 2>/dev/null
 	-$(INSTALL) -g @GID@ -o @UID@ -m $(INSTMODE_BIN) tools/genssl $(BINPATH)/inspircd-genssl 2>/dev/null
 	-$(INSTALL) -g @GID@ -o @UID@ -m $(INSTMODE_TXT) docs/conf/*.example $(EXAPATH)
+	-$(INSTALL) -g @GID@ -o @UID@ -m $(INSTMODE_TXT) docs/conf/codepages/*.example $(EXAPATH)/codepages
 	-$(INSTALL) -g @GID@ -o @UID@ -m $(INSTMODE_TXT) docs/conf/providers/*.example $(EXAPATH)/providers
 	-$(INSTALL) -g @GID@ -o @UID@ -m $(INSTMODE_TXT) docs/conf/services/*.example $(EXAPATH)/services
 	-$(INSTALL) -g @GID@ -o @UID@ -m $(INSTMODE_TXT) docs/sql/*.sql $(EXAPATH)/sql
diff --git a/src/modules/m_codepage.cpp b/src/modules/m_codepage.cpp
new file mode 100644
index 000000000..1c3ac02da
--- /dev/null
+++ b/src/modules/m_codepage.cpp
@@ -0,0 +1,212 @@
+/*
+ * InspIRCd -- Internet Relay Chat Daemon
+ *
+ *   Copyright (C) 2020 Sadie Powell <sadie@witchery.services>
+ *   Copyright (C) 2014 Googolplexed <googol@googolplexed.net>
+ *
+ * This file is part of InspIRCd.  InspIRCd is free software: you can
+ * redistribute it and/or modify it under the terms of the GNU General Public
+ * License as published by the Free Software Foundation, version 2.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+ * FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+
+#include "inspircd.h"
+
+typedef std::bitset<UCHAR_MAX + 1> AllowedChars;
+
+namespace
+{
+	// The characters which are allowed in nicknames.
+	AllowedChars allowedchars;
+
+	// The characters which are allowed at the front of a nickname.
+	AllowedChars allowedfrontchars;
+
+	// The mapping of lower case characters to upper case characters.
+	unsigned char casemap[UCHAR_MAX];
+
+	bool IsValidNick(const std::string& nick)
+	{
+		if (nick.empty() || nick.length() > ServerInstance->Config->Limits.NickMax)
+			return false;
+
+		for (std::string::const_iterator iter = nick.begin(); iter != nick.end(); ++iter)
+		{
+			unsigned char chr = static_cast<unsigned char>(*iter);
+
+			// Check that the character is allowed at the front of the nick.
+			if (iter == nick.begin() && !allowedfrontchars[chr])
+				return false;
+
+			// Check that the character is allowed in the nick.
+			if (!allowedchars[chr])
+				return false;
+		}
+
+		return true;
+	}
+}
+
+class ModuleCodepage
+	: public Module
+{
+ private:
+	// The character map which was set before this module was loaded.
+	const unsigned char* origcasemap;
+
+	// The IsNick handler which was set before this module was loaded.
+	TR1NS::function<bool(const std::string&)> origisnick;
+
+	template <typename T>
+	void RehashHashmap(T& hashmap)
+	{
+		T newhash(hashmap.bucket_count());
+		for (typename T::const_iterator i = hashmap.begin(); i != hashmap.end(); ++i)
+			newhash.insert(std::make_pair(i->first, i->second));
+		hashmap.swap(newhash);
+	}
+
+	void CheckInvalidNick()
+	{
+		const UserManager::LocalList& list = ServerInstance->Users.GetLocalUsers();
+		for (UserManager::LocalList::const_iterator iter = list.begin(); iter != list.end(); ++iter)
+		{
+			LocalUser* user = *iter;
+			if (user->nick != user->uuid && !ServerInstance->IsNick(user->nick))
+				user->ChangeNick(user->uuid);
+		}
+	}
+
+	void CheckRehash(unsigned char* prevmap)
+	{
+		if (!memcmp(prevmap, national_case_insensitive_map, sizeof(origcasemap)))
+			return;
+
+		RehashHashmap(ServerInstance->Users.clientlist);
+		RehashHashmap(ServerInstance->Users.uuidlist);
+		RehashHashmap(ServerInstance->chanlist);
+	}
+
+ public:
+	ModuleCodepage()
+		: origcasemap(national_case_insensitive_map)
+		, origisnick(ServerInstance->IsNick)
+	{
+	}
+
+	~ModuleCodepage()
+	{
+		ServerInstance->IsNick = origisnick;
+		CheckInvalidNick();
+
+		national_case_insensitive_map = origcasemap;
+		CheckRehash(casemap);
+	}
+
+	void ReadConfig(ConfigStatus& status) CXX11_OVERRIDE
+	{
+		const std::string name = ServerInstance->Config->ConfValue("codepage")->getString("name");
+		if (name.empty())
+			throw ModuleException("<codepage:name> is a required field!");
+
+		AllowedChars newallowedchars;
+		AllowedChars newallowedfrontchars;
+		ConfigTagList cpchars = ServerInstance->Config->ConfTags("cpchars");
+		for (ConfigIter i = cpchars.first; i != cpchars.second; ++i)
+		{
+			ConfigTag* tag = i->second;
+
+			unsigned char begin = tag->getUInt("begin", tag->getUInt("index", 0), 1, UCHAR_MAX);
+			if (!begin)
+				throw ModuleException("<cpchars> tag without index or begin specified at " + tag->getTagLocation());
+	
+			unsigned char end = tag->getUInt("end", begin, 1, UCHAR_MAX);
+			if (begin > end)
+				throw ModuleException("<cpchars:begin> must be lower than <cpchars:end> at " + tag->getTagLocation());
+
+			bool front = tag->getBool("front", false);
+			for (unsigned short pos = begin; pos <= end; ++pos)
+			{
+				if (pos == '\n' || pos == '\r' || pos == ' ')
+				{
+					throw ModuleException(InspIRCd::Format("<cpchars> tag contains a forbidden character: %u at %s",
+						pos, tag->getTagLocation().c_str()));
+				}
+
+				if (front && (pos == ':' || isdigit(pos)))
+				{
+					throw ModuleException(InspIRCd::Format("<cpchars> tag contains a forbidden front character: %u at %s",
+						pos, tag->getTagLocation().c_str()));
+				}
+
+				newallowedchars.set(pos);
+				newallowedfrontchars.set(pos, front);
+				ServerInstance->Logs->Log(MODNAME, LOG_DEBUG, "Marked %u (%c) as allowed (front: %s)",
+					pos, pos, front ? "yes" : "no");
+			}
+		}
+
+		unsigned char newcasemap[UCHAR_MAX];
+		for (size_t i = 0; i < UCHAR_MAX; ++i)
+			newcasemap[i] = i;
+		ConfigTagList cpcase = ServerInstance->Config->ConfTags("cpcase");
+		for (ConfigIter i = cpcase.first; i != cpcase.second; ++i)
+		{
+			ConfigTag* tag = i->second;
+
+			unsigned char lower = tag->getUInt("lower", 0, 1, UCHAR_MAX);
+			if (!lower)
+				throw ModuleException("<cpcase:lower> is required at " + tag->getTagLocation());
+
+			unsigned char upper = tag->getUInt("upper", 0, 1, UCHAR_MAX);
+			if (!upper)
+				throw ModuleException("<cpcase:upper> is required at " + tag->getTagLocation());
+
+			newcasemap[upper] = lower;
+			ServerInstance->Logs->Log(MODNAME, LOG_DEBUG, "Marked %u (%c) as the lower case version of %u (%c)",
+				lower, lower, upper, upper);
+		}
+
+		std::swap(allowedchars, newallowedchars);
+		std::swap(allowedfrontchars, newallowedfrontchars);
+		std::swap(casemap, newcasemap);
+
+		ServerInstance->IsNick = &IsValidNick;
+		CheckInvalidNick();
+
+		ServerInstance->Config->CaseMapping = name;
+		national_case_insensitive_map = casemap;
+		CheckRehash(newcasemap);
+	}
+
+	Version GetVersion() CXX11_OVERRIDE
+	{
+		std::stringstream linkdata;
+
+		linkdata << "front=";
+		for (size_t i = 0; i < allowedfrontchars.size(); ++i)
+			if (allowedfrontchars[i])
+				linkdata << static_cast<unsigned char>(i);
+
+		linkdata << "&middle=";
+		for (size_t i = 0; i < allowedchars.size(); ++i)
+			if (allowedchars[i])
+				linkdata << static_cast<unsigned char>(i);
+
+		linkdata << "&map=";
+		for (size_t i = 0; i < sizeof(casemap); ++i)
+			if (casemap[i] != i)
+				linkdata << static_cast<unsigned char>(i) << casemap[i] << ',';
+
+		return Version("Provides support for custom 8-bit codepages", VF_COMMON | VF_VENDOR, linkdata.str());
+	}
+};
+MODULE_INIT(ModuleCodepage)