From d84f6e012ca726cd37fb1b1fe8131620ba2021a2 Mon Sep 17 00:00:00 2001 From: Jon Siwek Date: Mon, 29 Jul 2013 15:41:34 -0500 Subject: [PATCH 1/6] Fix various documentation, mostly related to file analysis. - Fix examples/references in the file analysis how-to/usage doc. - Add Broxygen-generated docs for file analyzer plugins. - Break FTP::Info type declaration out in to its own file to get rid of some circular dependencies (between s/b/p/ftp/main and s/b/p/ftp/utils). --- doc/file-analysis.rst | 27 +++---- doc/index.rst | 1 + doc/scripts/CMakeLists.txt | 50 +++++++------ doc/scripts/DocSourcesList.cmake | 1 + scripts/base/frameworks/files/main.bro | 2 +- .../base/frameworks/packet-filter/main.bro | 2 +- scripts/base/init-bare.bro | 17 ++--- scripts/base/protocols/ftp/__load__.bro | 1 + scripts/base/protocols/ftp/files.bro | 1 + scripts/base/protocols/ftp/gridftp.bro | 1 + scripts/base/protocols/ftp/info.bro | 72 +++++++++++++++++++ scripts/base/protocols/ftp/main.bro | 67 +---------------- scripts/base/protocols/ftp/utils.bro | 5 +- .../policy/frameworks/packet-filter/shunt.bro | 4 +- scripts/policy/misc/load-balancing.bro | 12 ++-- .../tuning/defaults/packet-fragments.bro | 12 ++-- src/BroDoc.cc | 65 +++++++++++++++-- src/BroDoc.h | 6 ++ src/file_analysis/Manager.cc | 5 ++ src/file_analysis/Manager.h | 6 ++ src/main.cc | 1 + .../canonified_loaded_scripts.log | 7 +- 22 files changed, 227 insertions(+), 138 deletions(-) create mode 100644 scripts/base/protocols/ftp/info.bro diff --git a/doc/file-analysis.rst b/doc/file-analysis.rst index f312e06471..0a96a8efb7 100644 --- a/doc/file-analysis.rst +++ b/doc/file-analysis.rst @@ -82,9 +82,9 @@ attached, they start receiving the contents of the file as Bro extracts it from an ongoing network connection. What they do with the file contents is up to the particular file analyzer implementation, but they'll typically either report further information about the file via -events (e.g. :bro:see:`FileAnalysis::ANALYZER_MD5` will report the +events (e.g. :bro:see:`Files::ANALYZER_MD5` will report the file's MD5 checksum via :bro:see:`file_hash` once calculated) or they'll -have some side effect (e.g. :bro:see:`FileAnalysis::ANALYZER_EXTRACT` +have some side effect (e.g. :bro:see:`Files::ANALYZER_EXTRACT` will write the contents of the file out to the local file system). In the future there may be file analyzers that automatically attach to @@ -98,7 +98,7 @@ explicit attachment decision: { print "new file", f$id; if ( f?$mime_type && f$mime_type == "text/plain" ) - FileAnalysis::add_analyzer(f, [$tag=FileAnalysis::ANALYZER_MD5]); + Files::add_analyzer(f, Files::ANALYZER_MD5); } event file_hash(f: fa_file, kind: string, hash: string) @@ -113,26 +113,27 @@ output:: file_hash, Cx92a0ym5R8, md5, 397168fd09991a0e712254df7bc639ac Some file analyzers might have tunable parameters that need to be -specified in the call to :bro:see:`FileAnalysis::add_analyzer`: +specified in the call to :bro:see:`Files::add_analyzer`: .. code:: bro event file_new(f: fa_file) { - FileAnalysis::add_analyzer(f, [$tag=FileAnalysis::ANALYZER_EXTRACT, - $extract_filename="./myfile"]); + Files::add_analyzer(f, Files::ANALYZER_EXTRACT, + [$extract_filename="myfile"]); } In this case, the file extraction analyzer doesn't generate any further -events, but does have the side effect of writing out the file contents -to the local file system at the specified location of ``./myfile``. Of -course, for a network with more than a single file being transferred, -it's probably preferable to specify a different extraction path for each -file, unlike this example. +events, but does have the effect of writing out the file contents to the +local file system at the location resulting from the concatenation of +the path specified by :bro:see:`FileExtract::prefix` and the string, +``myfile``. Of course, for a network with more than a single file being +transferred, it's probably preferable to specify a different extraction +path for each file, unlike this example. Regardless of which file analyzers end up acting on a file, general information about the file (e.g. size, time of last data transferred, -MIME type, etc.) are logged in ``file_analysis.log``. +MIME type, etc.) are logged in ``files.log``. Input Framework Integration =========================== @@ -150,7 +151,7 @@ a network interface it's monitoring. It only requires a call to event file_new(f: fa_file) { print "new file", f$id; - FileAnalysis::add_analyzer(f, [$tag=FileAnalysis::ANALYZER_MD5]); + Files::add_analyzer(f, Files::ANALYZER_MD5); } event file_state_remove(f: fa_file) diff --git a/doc/index.rst b/doc/index.rst index ad05f7bf82..aa33d8797d 100644 --- a/doc/index.rst +++ b/doc/index.rst @@ -47,6 +47,7 @@ Script Reference scripts/index scripts/builtins scripts/proto-analyzers + scripts/file-analyzers Other Bro Components -------------------- diff --git a/doc/scripts/CMakeLists.txt b/doc/scripts/CMakeLists.txt index e7e39d0b3f..fa234e74f2 100644 --- a/doc/scripts/CMakeLists.txt +++ b/doc/scripts/CMakeLists.txt @@ -124,28 +124,34 @@ endmacro(REST_TARGET) # Schedule Bro scripts for which to generate documentation. include(DocSourcesList.cmake) -# This reST target is independent of a particular Bro script... -add_custom_command(OUTPUT proto-analyzers.rst - # delete any leftover state from previous bro runs - COMMAND "${CMAKE_COMMAND}" - ARGS -E remove_directory .state - # generate the reST documentation using bro - COMMAND BROPATH=${BROPATH}:${srcDir} BROMAGIC=${CMAKE_SOURCE_DIR}/magic/database ${CMAKE_BINARY_DIR}/src/bro - ARGS -b -Z base/init-bare.bro || (rm -rf .state *.log *.rst && exit 1) - # move generated doc into a new directory tree that - # defines the final structure of documents - COMMAND "${CMAKE_COMMAND}" - ARGS -E make_directory ${dstDir} - COMMAND "${CMAKE_COMMAND}" - ARGS -E copy proto-analyzers.rst ${dstDir} - # clean up the build directory - COMMAND rm - ARGS -rf .state *.log *.rst - DEPENDS bro - WORKING_DIRECTORY ${CMAKE_BINARY_DIR} - COMMENT "[Bro] Generating reST docs for proto-analyzers.rst" -) -list(APPEND ALL_REST_OUTPUTS proto-analyzers.rst) +# Macro for generating reST docs that are independent of any particular Bro +# script. +macro(INDEPENDENT_REST_TARGET reST_file) + add_custom_command(OUTPUT ${reST_file} + # delete any leftover state from previous bro runs + COMMAND "${CMAKE_COMMAND}" + ARGS -E remove_directory .state + # generate the reST documentation using bro + COMMAND BROPATH=${BROPATH}:${srcDir} BROMAGIC=${CMAKE_SOURCE_DIR}/magic/database ${CMAKE_BINARY_DIR}/src/bro + ARGS -b -Z base/init-bare.bro || (rm -rf .state *.log *.rst && exit 1) + # move generated doc into a new directory tree that + # defines the final structure of documents + COMMAND "${CMAKE_COMMAND}" + ARGS -E make_directory ${dstDir} + COMMAND "${CMAKE_COMMAND}" + ARGS -E copy ${reST_file} ${dstDir} + # clean up the build directory + COMMAND rm + ARGS -rf .state *.log *.rst + DEPENDS bro + WORKING_DIRECTORY ${CMAKE_BINARY_DIR} + COMMENT "[Bro] Generating reST docs for ${reST_file}" + ) + list(APPEND ALL_REST_OUTPUTS ${reST_file}) +endmacro(INDEPENDENT_REST_TARGET) + +independent_rest_target(proto-analyzers.rst) +independent_rest_target(file-analyzers.rst) # create temporary list of all docs to include in the master policy/index file file(WRITE ${MASTER_POLICY_INDEX} "${MASTER_POLICY_INDEX_TEXT}") diff --git a/doc/scripts/DocSourcesList.cmake b/doc/scripts/DocSourcesList.cmake index b2c932d117..d61db99db1 100644 --- a/doc/scripts/DocSourcesList.cmake +++ b/doc/scripts/DocSourcesList.cmake @@ -140,6 +140,7 @@ rest_target(${psd} base/protocols/dns/consts.bro) rest_target(${psd} base/protocols/dns/main.bro) rest_target(${psd} base/protocols/ftp/files.bro) rest_target(${psd} base/protocols/ftp/gridftp.bro) +rest_target(${psd} base/protocols/ftp/info.bro) rest_target(${psd} base/protocols/ftp/main.bro) rest_target(${psd} base/protocols/ftp/utils-commands.bro) rest_target(${psd} base/protocols/ftp/utils.bro) diff --git a/scripts/base/frameworks/files/main.bro b/scripts/base/frameworks/files/main.bro index d0c381545b..a87608054d 100644 --- a/scripts/base/frameworks/files/main.bro +++ b/scripts/base/frameworks/files/main.bro @@ -204,7 +204,7 @@ export { ## ## tag: Tag for the protocol analyzer having a callback being registered. ## - ## reg: A :bro:see:`ProtoRegistration` record. + ## reg: A :bro:see:`Files::ProtoRegistration` record. ## ## Returns: true if the protocol being registered was not previously registered. global register_protocol: function(tag: Analyzer::Tag, reg: ProtoRegistration): bool; diff --git a/scripts/base/frameworks/packet-filter/main.bro b/scripts/base/frameworks/packet-filter/main.bro index 72b2b62f34..929b10fbe1 100644 --- a/scripts/base/frameworks/packet-filter/main.bro +++ b/scripts/base/frameworks/packet-filter/main.bro @@ -109,7 +109,7 @@ export { ## Enables the old filtering approach of "only watch common ports for ## analyzed protocols". - ## + ## ## Unless you know what you are doing, leave this set to F. const enable_auto_protocol_capture_filters = F &redef; diff --git a/scripts/base/init-bare.bro b/scripts/base/init-bare.bro index 594991c85a..92b806092c 100644 --- a/scripts/base/init-bare.bro +++ b/scripts/base/init-bare.bro @@ -531,22 +531,19 @@ type record_field_table: table[string] of record_field; # dependent on the names remaining as they are now. ## Set of BPF capture filters to use for capturing, indexed by a user-definable -## ID (which must be unique). If Bro is *not* configured to examine -## :bro:id:`PacketFilter::all_packets`, all packets matching at least -## one of the filters in this table (and all in :bro:id:`restrict_filters`) -## will be analyzed. +## ID (which must be unique). If Bro is *not* configured with +## :bro:id:`PacketFilter::enable_auto_protocol_capture_filters`, +## all packets matching at least one of the filters in this table (and all in +## :bro:id:`restrict_filters`) will be analyzed. ## -## .. bro:see:: PacketFilter PacketFilter::all_packets +## .. bro:see:: PacketFilter PacketFilter::enable_auto_protocol_capture_filters ## PacketFilter::unrestricted_filter restrict_filters global capture_filters: table[string] of string &redef; ## Set of BPF filters to restrict capturing, indexed by a user-definable ID (which -## must be unique). If Bro is *not* configured to examine -## :bro:id:`PacketFilter::all_packets`, only packets matching *all* of the -## filters in this table (and any in :bro:id:`capture_filters`) will be -## analyzed. +## must be unique). ## -## .. bro:see:: PacketFilter PacketFilter::all_packets +## .. bro:see:: PacketFilter PacketFilter::enable_auto_protocol_capture_filters ## PacketFilter::unrestricted_filter capture_filters global restrict_filters: table[string] of string &redef; diff --git a/scripts/base/protocols/ftp/__load__.bro b/scripts/base/protocols/ftp/__load__.bro index ebb09e702c..3ddd8a2dc2 100644 --- a/scripts/base/protocols/ftp/__load__.bro +++ b/scripts/base/protocols/ftp/__load__.bro @@ -1,4 +1,5 @@ @load ./utils-commands +@load ./info @load ./main @load ./utils @load ./files diff --git a/scripts/base/protocols/ftp/files.bro b/scripts/base/protocols/ftp/files.bro index 9ed17ab2a4..b507ca32a7 100644 --- a/scripts/base/protocols/ftp/files.bro +++ b/scripts/base/protocols/ftp/files.bro @@ -1,3 +1,4 @@ +@load ./info @load ./main @load ./utils @load base/utils/conn-ids diff --git a/scripts/base/protocols/ftp/gridftp.bro b/scripts/base/protocols/ftp/gridftp.bro index 57752b1cbd..73bd656544 100644 --- a/scripts/base/protocols/ftp/gridftp.bro +++ b/scripts/base/protocols/ftp/gridftp.bro @@ -19,6 +19,7 @@ ##! sizes are not logged, but at the benefit of saving CPU cycles that ##! otherwise go to analyzing the large (and likely benign) connections. +@load ./info @load ./main @load base/protocols/conn @load base/protocols/ssl diff --git a/scripts/base/protocols/ftp/info.bro b/scripts/base/protocols/ftp/info.bro new file mode 100644 index 0000000000..f6fceb071e --- /dev/null +++ b/scripts/base/protocols/ftp/info.bro @@ -0,0 +1,72 @@ +##! Defines data structures for tracking and logging FTP sessions. + +module FTP; + +@load ./utils-commands + +export { + + ## This setting changes if passwords used in FTP sessions are + ## captured or not. + const default_capture_password = F &redef; + + ## The expected endpoints of an FTP data channel. + type ExpectedDataChannel: record { + ## Whether PASV mode is toggled for control channel. + passive: bool &log; + ## The host that will be initiating the data connection. + orig_h: addr &log; + ## The host that will be accepting the data connection. + resp_h: addr &log; + ## The port at which the acceptor is listening for the data connection. + resp_p: port &log; + }; + + type Info: record { + ## Time when the command was sent. + ts: time &log; + ## Unique ID for the connection. + uid: string &log; + ## The connection's 4-tuple of endpoint addresses/ports. + id: conn_id &log; + ## User name for the current FTP session. + user: string &log &default=""; + ## Password for the current FTP session if captured. + password: string &log &optional; + ## Command given by the client. + command: string &log &optional; + ## Argument for the command if one is given. + arg: string &log &optional; + + ## Libmagic "sniffed" file type if the command indicates a file transfer. + mime_type: string &log &optional; + ## Size of the file if the command indicates a file transfer. + file_size: count &log &optional; + + ## Reply code from the server in response to the command. + reply_code: count &log &optional; + ## Reply message from the server in response to the command. + reply_msg: string &log &optional; + + ## Expected FTP data channel. + data_channel: ExpectedDataChannel &log &optional; + + ## Current working directory that this session is in. By making + ## the default value '.', we can indicate that unless something + ## more concrete is discovered that the existing but unknown + ## directory is ok to use. + cwd: string &default="."; + + ## Command that is currently waiting for a response. + cmdarg: CmdArg &optional; + ## Queue for commands that have been sent but not yet responded to + ## are tracked here. + pending_commands: PendingCmds; + + ## Indicates if the session is in active or passive mode. + passive: bool &default=F; + + ## Determines if the password will be captured for this request. + capture_password: bool &default=default_capture_password; + }; +} diff --git a/scripts/base/protocols/ftp/main.bro b/scripts/base/protocols/ftp/main.bro index c9549a14ec..df66235d49 100644 --- a/scripts/base/protocols/ftp/main.bro +++ b/scripts/base/protocols/ftp/main.bro @@ -3,6 +3,8 @@ ##! will take on the full path that the client is at along with the requested ##! file name. +@load ./info +@load ./utils @load ./utils-commands @load base/utils/paths @load base/utils/numbers @@ -20,72 +22,9 @@ export { "EPSV" } &redef; - ## This setting changes if passwords used in FTP sessions are captured or not. - const default_capture_password = F &redef; - ## User IDs that can be considered "anonymous". const guest_ids = { "anonymous", "ftp", "ftpuser", "guest" } &redef; - ## The expected endpoints of an FTP data channel. - type ExpectedDataChannel: record { - ## Whether PASV mode is toggled for control channel. - passive: bool &log; - ## The host that will be initiating the data connection. - orig_h: addr &log; - ## The host that will be accepting the data connection. - resp_h: addr &log; - ## The port at which the acceptor is listening for the data connection. - resp_p: port &log; - }; - - type Info: record { - ## Time when the command was sent. - ts: time &log; - ## Unique ID for the connection. - uid: string &log; - ## The connection's 4-tuple of endpoint addresses/ports. - id: conn_id &log; - ## User name for the current FTP session. - user: string &log &default=""; - ## Password for the current FTP session if captured. - password: string &log &optional; - ## Command given by the client. - command: string &log &optional; - ## Argument for the command if one is given. - arg: string &log &optional; - - ## Libmagic "sniffed" file type if the command indicates a file transfer. - mime_type: string &log &optional; - ## Size of the file if the command indicates a file transfer. - file_size: count &log &optional; - - ## Reply code from the server in response to the command. - reply_code: count &log &optional; - ## Reply message from the server in response to the command. - reply_msg: string &log &optional; - - ## Expected FTP data channel. - data_channel: ExpectedDataChannel &log &optional; - - ## Current working directory that this session is in. By making - ## the default value '.', we can indicate that unless something - ## more concrete is discovered that the existing but unknown - ## directory is ok to use. - cwd: string &default="."; - - ## Command that is currently waiting for a response. - cmdarg: CmdArg &optional; - ## Queue for commands that have been sent but not yet responded to - ## are tracked here. - pending_commands: PendingCmds; - - ## Indicates if the session is in active or passive mode. - passive: bool &default=F; - - ## Determines if the password will be captured for this request. - capture_password: bool &default=default_capture_password; - }; - ## This record is to hold a parsed FTP reply code. For example, for the ## 201 status code, the digits would be parsed as: x->2, y->0, z=>1. type ReplyCode: record { @@ -102,8 +41,6 @@ export { global log_ftp: event(rec: Info); } -@load ./utils - # Add the state tracking information variable to the connection record redef record connection += { ftp: Info &optional; diff --git a/scripts/base/protocols/ftp/utils.bro b/scripts/base/protocols/ftp/utils.bro index 629b87e5a8..a0b473e086 100644 --- a/scripts/base/protocols/ftp/utils.bro +++ b/scripts/base/protocols/ftp/utils.bro @@ -1,7 +1,8 @@ ##! Utilities specific for FTP processing. -@load ./main +@load ./info @load base/utils/addrs +@load base/utils/paths module FTP; @@ -44,4 +45,4 @@ function build_url_ftp(rec: Info): string function describe(rec: Info): string { return build_url_ftp(rec); - } \ No newline at end of file + } diff --git a/scripts/policy/frameworks/packet-filter/shunt.bro b/scripts/policy/frameworks/packet-filter/shunt.bro index b87369ee62..85ec189a17 100644 --- a/scripts/policy/frameworks/packet-filter/shunt.bro +++ b/scripts/policy/frameworks/packet-filter/shunt.bro @@ -34,8 +34,8 @@ export { global current_shunted_host_pairs: function(): set[conn_id]; redef enum Notice::Type += { - ## Indicative that :bro:id:`max_bpf_shunts` connections are already - ## being shunted with BPF filters and no more are allowed. + ## Indicative that :bro:id:`PacketFilter::max_bpf_shunts` connections + ## are already being shunted with BPF filters and no more are allowed. No_More_Conn_Shunts_Available, ## Limitations in BPF make shunting some connections with BPF impossible. diff --git a/scripts/policy/misc/load-balancing.bro b/scripts/policy/misc/load-balancing.bro index fe07dd64da..889d18119a 100644 --- a/scripts/policy/misc/load-balancing.bro +++ b/scripts/policy/misc/load-balancing.bro @@ -12,12 +12,12 @@ export { ## Apply BPF filters to each worker in a way that causes them to ## automatically flow balance traffic between them. AUTO_BPF, - ## Load balance traffic across the workers by making each one apply - ## a restrict filter to only listen to a single MAC address. This - ## is a somewhat common deployment option for sites doing network - ## based load balancing with MAC address rewriting and passing the - ## traffic to a single interface. Multiple MAC addresses will show - ## up on the same interface and need filtered to a single address. + # Load balance traffic across the workers by making each one apply + # a restrict filter to only listen to a single MAC address. This + # is a somewhat common deployment option for sites doing network + # based load balancing with MAC address rewriting and passing the + # traffic to a single interface. Multiple MAC addresses will show + # up on the same interface and need filtered to a single address. #MAC_ADDR_BPF, }; diff --git a/scripts/policy/tuning/defaults/packet-fragments.bro b/scripts/policy/tuning/defaults/packet-fragments.bro index 24b18d5917..f95c826547 100644 --- a/scripts/policy/tuning/defaults/packet-fragments.bro +++ b/scripts/policy/tuning/defaults/packet-fragments.bro @@ -1,10 +1,10 @@ -## Capture TCP fragments, but not UDP (or ICMP), since those are a lot more -## common due to high-volume, fragmenting protocols such as NFS :-(. +# Capture TCP fragments, but not UDP (or ICMP), since those are a lot more +# common due to high-volume, fragmenting protocols such as NFS :-(. -## This normally isn't used because of the default open packet filter -## but we set it anyway in case the user is using a packet filter. -## Note: This was removed because the default model now is to have a wide -## open packet filter. +# This normally isn't used because of the default open packet filter +# but we set it anyway in case the user is using a packet filter. +# Note: This was removed because the default model now is to have a wide +# open packet filter. #redef capture_filters += { ["frag"] = "(ip[6:2] & 0x3fff != 0) and tcp" }; ## Shorten the fragment timeout from never expiring to expiring fragments after diff --git a/src/BroDoc.cc b/src/BroDoc.cc index c04cd92eca..3cb271bdbf 100644 --- a/src/BroDoc.cc +++ b/src/BroDoc.cc @@ -11,6 +11,7 @@ #include "plugin/Manager.h" #include "analyzer/Manager.h" #include "analyzer/Component.h" +#include "file_analysis/Manager.h" BroDoc::BroDoc(const std::string& rel, const std::string& abs) { @@ -479,6 +480,17 @@ static void WriteAnalyzerComponent(FILE* f, const analyzer::Component* c) fprintf(f, ":bro:enum:`Analyzer::%s`\n\n", tag.c_str()); } +static void WriteAnalyzerComponent(FILE* f, const file_analysis::Component* c) + { + EnumType* atag = file_mgr->GetTagEnumType(); + string tag = fmt("ANALYZER_%s", c->CanonicalName()); + + if ( atag->Lookup("Files", tag.c_str()) < 0 ) + reporter->InternalError("missing analyzer tag for %s", tag.c_str()); + + fprintf(f, ":bro:enum:`Files::%s`\n\n", tag.c_str()); + } + static void WritePluginComponents(FILE* f, const plugin::Plugin* p) { plugin::Plugin::component_list components = p->Components(); @@ -494,6 +506,10 @@ static void WritePluginComponents(FILE* f, const plugin::Plugin* p) WriteAnalyzerComponent(f, dynamic_cast(*it)); break; + case plugin::component::FILE_ANALYZER: + WriteAnalyzerComponent(f, + dynamic_cast(*it)); + break; case plugin::component::READER: reporter->InternalError("docs for READER component unimplemented"); case plugin::component::WRITER: @@ -537,12 +553,13 @@ static void WritePluginBifItems(FILE* f, const plugin::Plugin* p, } } -static void WriteAnalyzerTagDefn(FILE* f, EnumType* e) +static void WriteAnalyzerTagDefn(FILE* f, EnumType* e, const string& module) { + string tag_id= module + "::Tag"; e = new CommentedEnumType(e); - e->SetTypeID(copy_string("Analyzer::Tag")); + e->SetTypeID(copy_string(tag_id.c_str())); - ID* dummy_id = new ID(copy_string("Analyzer::Tag"), SCOPE_GLOBAL, true); + ID* dummy_id = new ID(copy_string(tag_id.c_str()), SCOPE_GLOBAL, true); dummy_id->SetType(e); dummy_id->MakeType(); @@ -554,13 +571,17 @@ static void WriteAnalyzerTagDefn(FILE* f, EnumType* e) bdo.WriteReST(f); } -static bool IsAnalyzerPlugin(const plugin::Plugin* p) +static bool ComponentsMatch(const plugin::Plugin* p, plugin::component::Type t, + bool match_empty = false) { plugin::Plugin::component_list components = p->Components(); plugin::Plugin::component_list::const_iterator it; + if ( components.empty() ) + return match_empty; + for ( it = components.begin(); it != components.end(); ++it ) - if ( (*it)->Type() != plugin::component::ANALYZER ) + if ( (*it)->Type() != t ) return false; return true; @@ -573,14 +594,44 @@ void CreateProtoAnalyzerDoc(const char* filename) fprintf(f, "Protocol Analyzer Reference\n"); fprintf(f, "===========================\n\n"); - WriteAnalyzerTagDefn(f, analyzer_mgr->GetTagEnumType()); + WriteAnalyzerTagDefn(f, analyzer_mgr->GetTagEnumType(), "Analyzer"); plugin::Manager::plugin_list plugins = plugin_mgr->Plugins(); plugin::Manager::plugin_list::const_iterator it; for ( it = plugins.begin(); it != plugins.end(); ++it ) { - if ( ! IsAnalyzerPlugin(*it) ) + if ( ! ComponentsMatch(*it, plugin::component::ANALYZER, true) ) + continue; + + WritePluginSectionHeading(f, *it); + WritePluginComponents(f, *it); + WritePluginBifItems(f, *it, plugin::BifItem::CONSTANT, + "Options/Constants"); + WritePluginBifItems(f, *it, plugin::BifItem::GLOBAL, "Globals"); + WritePluginBifItems(f, *it, plugin::BifItem::TYPE, "Types"); + WritePluginBifItems(f, *it, plugin::BifItem::EVENT, "Events"); + WritePluginBifItems(f, *it, plugin::BifItem::FUNCTION, "Functions"); + } + + fclose(f); + } + +void CreateFileAnalyzerDoc(const char* filename) + { + FILE* f = fopen(filename, "w"); + + fprintf(f, "File Analyzer Reference\n"); + fprintf(f, "===========================\n\n"); + + WriteAnalyzerTagDefn(f, file_mgr->GetTagEnumType(), "Files"); + + plugin::Manager::plugin_list plugins = plugin_mgr->Plugins(); + plugin::Manager::plugin_list::const_iterator it; + + for ( it = plugins.begin(); it != plugins.end(); ++it ) + { + if ( ! ComponentsMatch(*it, plugin::component::FILE_ANALYZER) ) continue; WritePluginSectionHeading(f, *it); diff --git a/src/BroDoc.h b/src/BroDoc.h index 9f92f821f8..081df698d9 100644 --- a/src/BroDoc.h +++ b/src/BroDoc.h @@ -413,4 +413,10 @@ private: */ void CreateProtoAnalyzerDoc(const char* filename); +/** + * Writes out plugin index documentation for all file analyzer plugins. + * @param filename the name of the file to write. + */ +void CreateFileAnalyzerDoc(const char* filename); + #endif diff --git a/src/file_analysis/Manager.cc b/src/file_analysis/Manager.cc index 4e25bb0b0e..fb74a409b4 100644 --- a/src/file_analysis/Manager.cc +++ b/src/file_analysis/Manager.cc @@ -394,3 +394,8 @@ const char* Manager::GetAnalyzerName(int tag) const return it->second->CanonicalName(); } + +EnumType* Manager::GetTagEnumType() + { + return tag_enum_type; + } diff --git a/src/file_analysis/Manager.h b/src/file_analysis/Manager.h index 84b606173d..a93e78c638 100644 --- a/src/file_analysis/Manager.h +++ b/src/file_analysis/Manager.h @@ -214,6 +214,12 @@ public: */ const char* GetAnalyzerName(int tag) const; + /** + * Returns the enum type that corresponds to the script-level type + * \c Files::Tag. + */ + EnumType* GetTagEnumType(); + protected: friend class FileTimer; diff --git a/src/main.cc b/src/main.cc index 56193a935b..6a58832964 100644 --- a/src/main.cc +++ b/src/main.cc @@ -872,6 +872,7 @@ int main(int argc, char** argv) if ( generate_documentation ) { CreateProtoAnalyzerDoc("proto-analyzers.rst"); + CreateFileAnalyzerDoc("file-analyzers.rst"); std::list::iterator it; diff --git a/testing/btest/Baseline/coverage.default-load-baseline/canonified_loaded_scripts.log b/testing/btest/Baseline/coverage.default-load-baseline/canonified_loaded_scripts.log index f67d4b6158..c34e2e2e87 100644 --- a/testing/btest/Baseline/coverage.default-load-baseline/canonified_loaded_scripts.log +++ b/testing/btest/Baseline/coverage.default-load-baseline/canonified_loaded_scripts.log @@ -3,7 +3,7 @@ #empty_field (empty) #unset_field - #path loaded_scripts -#open 2013-07-23-05-48-10 +#open 2013-07-29-20-08-38 #fields name #types string scripts/base/init-bare.bro @@ -156,8 +156,9 @@ scripts/base/init-default.bro scripts/base/protocols/dns/main.bro scripts/base/protocols/ftp/__load__.bro scripts/base/protocols/ftp/utils-commands.bro + scripts/base/protocols/ftp/info.bro scripts/base/protocols/ftp/main.bro - scripts/base/protocols/ftp/utils.bro + scripts/base/protocols/ftp/utils.bro scripts/base/protocols/ftp/files.bro scripts/base/protocols/ftp/gridftp.bro scripts/base/protocols/ssl/__load__.bro @@ -196,4 +197,4 @@ scripts/base/init-default.bro scripts/base/files/extract/main.bro scripts/base/misc/find-checksum-offloading.bro scripts/policy/misc/loaded-scripts.bro -#close 2013-07-23-05-48-10 +#close 2013-07-29-20-08-38 From 8df4df0b8b7c8760d830c0f99e26e8f4db66967a Mon Sep 17 00:00:00 2001 From: Jon Siwek Date: Tue, 30 Jul 2013 15:19:48 -0500 Subject: [PATCH 2/6] Add a distinct tag class for file analyzers. This should prevent assignment mismatches between file and protocol analyzer tags. --- src/CMakeLists.txt | 1 + src/Tag.cc | 82 ++++++++++++++++++ src/Tag.h | 138 +++++++++++++++++++++++++++++++ src/analyzer/Component.cc | 2 +- src/analyzer/Manager.cc | 2 +- src/analyzer/Tag.cc | 84 ++----------------- src/analyzer/Tag.h | 93 ++++++--------------- src/file_analysis/CMakeLists.txt | 1 + src/file_analysis/Component.cc | 8 +- src/file_analysis/Component.h | 16 ++-- src/file_analysis/Manager.h | 2 +- src/file_analysis/Tag.cc | 24 ++++++ src/file_analysis/Tag.h | 115 ++++++++++++++++++++++++++ 13 files changed, 407 insertions(+), 161 deletions(-) create mode 100644 src/Tag.cc create mode 100644 src/Tag.h create mode 100644 src/file_analysis/Tag.cc create mode 100644 src/file_analysis/Tag.h diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index e353dd4695..082f34fba1 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -315,6 +315,7 @@ set(bro_SRCS StateAccess.cc Stats.cc Stmt.cc + Tag.cc Timer.cc Traverse.cc Trigger.cc diff --git a/src/Tag.cc b/src/Tag.cc new file mode 100644 index 0000000000..178edaa71e --- /dev/null +++ b/src/Tag.cc @@ -0,0 +1,82 @@ +// See the file "COPYING" in the main distribution directory for copyright. + +#include "Tag.h" +#include "Val.h" + +Tag::Tag(EnumType* etype, type_t arg_type, subtype_t arg_subtype) + { + assert(arg_type > 0); + + type = arg_type; + subtype = arg_subtype; + int64_t i = (int64)(type) | ((int64)subtype << 31); + Ref(etype); + val = new EnumVal(i, etype); + } + +Tag::Tag(EnumVal* arg_val) + { + assert(arg_val); + + val = arg_val; + Ref(val); + + int64 i = val->InternalInt(); + type = i & 0xffffffff; + subtype = (i >> 31) & 0xffffffff; + } + +Tag::Tag(const Tag& other) + { + type = other.type; + subtype = other.subtype; + val = other.val; + + if ( val ) + Ref(val); + } + +Tag::Tag() + { + type = 0; + subtype = 0; + val = 0; + } + +Tag::~Tag() + { + Unref(val); + val = 0; + } + +Tag& Tag::operator=(const Tag& other) + { + if ( this != &other ) + { + type = other.type; + subtype = other.subtype; + val = other.val; + + if ( val ) + Ref(val); + } + + return *this; + } + +EnumVal* Tag::AsEnumVal(EnumType* etype) const + { + if ( ! val ) + { + assert(type == 0 && subtype == 0); + Ref(etype); + val = new EnumVal(0, etype); + } + + return val; + } + +std::string Tag::AsString() const + { + return fmt("%" PRIu32 "/%" PRIu32, type, subtype); + } diff --git a/src/Tag.h b/src/Tag.h new file mode 100644 index 0000000000..a0c218019e --- /dev/null +++ b/src/Tag.h @@ -0,0 +1,138 @@ +// See the file "COPYING" in the main distribution directory for copyright. + +#ifndef TAG_H +#define TAG_H + +#include "config.h" +#include "util.h" +#include "Type.h" + +class EnumVal; + +/** + * Class to identify an analyzer type. + * + * Each analyzer type gets a tag consisting of a main type and subtype. The + * former is an identifier that's unique all analyzer classes. The latter is + * passed through to the analyzer instances for their use, yet not further + * interpreted by the analyzer infrastructure; it allows an analyzer to + * branch out into a set of sub-analyzers internally. Jointly, main type and + * subtype form an analyzer "tag". Each unique tag corresponds to a single + * "analyzer" from the user's perspective. At the script layer, these tags + * are mapped into enums of type \c Analyzer::Tag or Files::Tag. Internally, + * the analyzer::Manager and file_analysis::Manager maintain the mapping of tag + * to analyzer (and it also assigns them their main types), and + * analyzer::Component and file_analysis::Component create new tag. + * + * The Tag class supports all operations necessary to act as an index in a + * \c std::map. + */ +class Tag { +public: + /** + * Type for the analyzer's main type. + */ + typedef uint32 type_t; + + /** + * Type for the analyzer's subtype. + */ + typedef uint32 subtype_t; + + /** + * Returns the tag's main type. + */ + type_t Type() const { return type; } + + /** + * Returns the tag's subtype. + */ + subtype_t Subtype() const { return subtype; } + + /** + * Returns the numerical values for main and subtype inside a string + * suitable for printing. This is primarily for debugging. + */ + std::string AsString() const; + +protected: + /* + * Copy constructor. + */ + Tag(const Tag& other); + + /** + * Default constructor. This initializes the tag with an error value + * that will make \c operator \c bool return false. + */ + Tag(); + + /** + * Destructor. + */ + ~Tag(); + + /** + * Assignment operator. + */ + Tag& operator=(const Tag& other); + + /** + * Compares two tags for equality. + */ + bool operator==(const Tag& other) const + { + return type == other.type && subtype == other.subtype; + } + + /** + * Compares two tags for inequality. + */ + bool operator!=(const Tag& other) const + { + return type != other.type || subtype != other.subtype; + } + + /** + * Compares two tags for less-than relationship. + */ + bool operator<(const Tag& other) const + { + return type != other.type ? type < other.type : (subtype < other.subtype); + } + + /** + * Returns the script-layer enum that corresponds to this tag. + * The returned value does not have its ref-count increased. + * + * @param etype the script-layer enum type associated with the tag. + */ + EnumVal* AsEnumVal(EnumType* etype) const; + + /** + * Constructor. + * + * @param etype the script-layer enum type associated with the tag. + * + * @param type The main type. Note that the manager class manages the + * the value space internally, so noone else should assign main types. + * + * @param subtype The sub type, which is left to an analyzer for + * interpretation. By default it's set to zero. + */ + Tag(EnumType* etype, type_t type, subtype_t subtype = 0); + + /** + * Constructor. + * + * @param val An enum value of script type \c Analyzer::Tag. + */ + Tag(EnumVal* val); + +private: + type_t type; // Main type. + subtype_t subtype; // Subtype. + mutable EnumVal* val; // Script-layer value. +}; + +#endif diff --git a/src/analyzer/Component.cc b/src/analyzer/Component.cc index cbb0f40c20..ded0a1a2d5 100644 --- a/src/analyzer/Component.cc +++ b/src/analyzer/Component.cc @@ -8,7 +8,7 @@ using namespace analyzer; -Tag::type_t Component::type_counter = 0; +analyzer::Tag::type_t Component::type_counter = 0; Component::Component(const char* arg_name, factory_callback arg_factory, Tag::subtype_t arg_subtype, bool arg_enabled, bool arg_partial) : plugin::Component(plugin::component::ANALYZER) diff --git a/src/analyzer/Manager.cc b/src/analyzer/Manager.cc index 5695dec625..3f97cbb0c8 100644 --- a/src/analyzer/Manager.cc +++ b/src/analyzer/Manager.cc @@ -341,7 +341,7 @@ const char* Manager::GetAnalyzerName(Val* val) return GetAnalyzerName(Tag(val->AsEnumVal())); } -Tag Manager::GetAnalyzerTag(const char* name) +analyzer::Tag Manager::GetAnalyzerTag(const char* name) { Component* c = Lookup(name); return c ? c->Tag() : Tag(); diff --git a/src/analyzer/Tag.cc b/src/analyzer/Tag.cc index 2f04ff17da..3ab41daf78 100644 --- a/src/analyzer/Tag.cc +++ b/src/analyzer/Tag.cc @@ -3,90 +3,20 @@ #include "Tag.h" #include "Manager.h" -#include "../NetVar.h" +analyzer::Tag analyzer::Tag::Error; -using namespace analyzer; - -Tag Tag::Error; - -Tag::Tag(type_t arg_type, subtype_t arg_subtype) +analyzer::Tag::Tag(type_t type, subtype_t subtype) + : ::Tag(analyzer_mgr->GetTagEnumType(), type, subtype) { - assert(arg_type > 0); - - type = arg_type; - subtype = arg_subtype; - int64_t i = (int64)(type) | ((int64)subtype << 31); - - EnumType* etype = analyzer_mgr->GetTagEnumType(); - Ref(etype); - val = new EnumVal(i, etype); } -Tag::Tag(EnumVal* arg_val) +analyzer::Tag& analyzer::Tag::operator=(const analyzer::Tag& other) { - assert(arg_val); - - val = arg_val; - Ref(val); - - int64 i = val->InternalInt(); - type = i & 0xffffffff; - subtype = (i >> 31) & 0xffffffff; - } - -Tag::Tag(const Tag& other) - { - type = other.type; - subtype = other.subtype; - val = other.val; - - if ( val ) - Ref(val); - } - -Tag::Tag() - { - type = 0; - subtype = 0; - val = 0; - } - -Tag::~Tag() - { - Unref(val); - val = 0; - } - -Tag& Tag::operator=(const Tag& other) - { - if ( this != &other ) - { - type = other.type; - subtype = other.subtype; - val = other.val; - - if ( val ) - Ref(val); - } - + ::Tag::operator=(other); return *this; } -EnumVal* Tag::AsEnumVal() const +EnumVal* analyzer::Tag::AsEnumVal() const { - if ( ! val ) - { - assert(analyzer_mgr); - assert(type == 0 && subtype == 0); - EnumType* etype = analyzer_mgr->GetTagEnumType(); - Ref(etype); - val = new EnumVal(0, etype); - } - - return val; - } - -std::string Tag::AsString() const - { - return fmt("%" PRIu32 "/%" PRIu32, type, subtype); + return ::Tag::AsEnumVal(analyzer_mgr->GetTagEnumType()); } diff --git a/src/analyzer/Tag.h b/src/analyzer/Tag.h index edb0ade8a7..8ac151e4b5 100644 --- a/src/analyzer/Tag.h +++ b/src/analyzer/Tag.h @@ -5,90 +5,44 @@ #include "config.h" #include "util.h" +#include "../Tag.h" class EnumVal; -namespace file_analysis { -class Manager; -class Component; -} - namespace analyzer { class Manager; class Component; /** - * Class to identify an analyzer type. + * Class to identify a protocol analyzer type. * - * Each analyzer type gets a tag consisting of a main type and subtype. The - * former is an identifier that's unique all analyzer classes. The latter is - * passed through to the analyzer instances for their use, yet not further - * interpreted by the analyzer infrastructure; it allows an analyzer to - * branch out into a set of sub-analyzers internally. Jointly, main type and - * subtype form an analyzer "tag". Each unique tag corresponds to a single - * "analyzer" from the user's perspective. At the script layer, these tags - * are mapped into enums of type \c Analyzer::Tag. Internally, the - * analyzer::Manager maintains the mapping of tag to analyzer (and it also - * assigns them their main types), and analyzer::Component creates new - * tags. - * - * The Tag class supports all operations necessary to act as an index in a - * \c std::map. + * The script-layer analogue is Analyzer::Tag. */ -class Tag { +class Tag : public ::Tag { public: - /** - * Type for the analyzer's main type. - */ - typedef uint32 type_t; - - /** - * Type for the analyzer's subtype. - */ - typedef uint32 subtype_t; - /* * Copy constructor. */ - Tag(const Tag& other); + Tag(const Tag& other) : ::Tag(other) {} /** * Default constructor. This initializes the tag with an error value * that will make \c operator \c bool return false. */ - Tag(); + Tag() : ::Tag() {} /** * Destructor. */ - ~Tag(); - - /** - * Returns the tag's main type. - */ - type_t Type() const { return type; } - - /** - * Returns the tag's subtype. - */ - subtype_t Subtype() const { return subtype; } - - /** - * Returns the \c Analyzer::Tag enum that corresponds to this tag. - * The returned value is \a does not have its ref-count increased. - */ - EnumVal* AsEnumVal() const; - - /** - * Returns the numerical values for main and subtype inside a string - * suitable for printing. This is primarily for debugging. - */ - std::string AsString() const; + ~Tag() {} /** * Returns false if the tag represents an error value rather than a * legal analyzer type. + * TODO: make this conversion operator "explicit" (C++11) or use a + * "safe bool" idiom (not necessary if "explicit" is available), + * otherwise this may allow nonsense/undesired comparison operations. */ operator bool() const { return *this != Tag(); } @@ -102,7 +56,7 @@ public: */ bool operator==(const Tag& other) const { - return type == other.type && subtype == other.subtype; + return ::Tag::operator==(other); } /** @@ -110,7 +64,7 @@ public: */ bool operator!=(const Tag& other) const { - return type != other.type || subtype != other.subtype; + return ::Tag::operator!=(other); } /** @@ -118,23 +72,29 @@ public: */ bool operator<(const Tag& other) const { - return type != other.type ? type < other.type : (subtype < other.subtype); + return ::Tag::operator<(other); } + /** + * Returns the \c Analyzer::Tag enum that corresponds to this tag. + * The returned value does not have its ref-count increased. + * + * @param etype the script-layer enum type associated with the tag. + */ + EnumVal* AsEnumVal() const; + static Tag Error; protected: friend class analyzer::Manager; friend class analyzer::Component; - friend class file_analysis::Manager; - friend class file_analysis::Component; /** * Constructor. * * @param type The main type. Note that the \a analyzer::Manager * manages the value space internally, so noone else should assign - * any main tyoes. + * any main types. * * @param subtype The sub type, which is left to an analyzer for * interpretation. By default it's set to zero. @@ -144,14 +104,9 @@ protected: /** * Constructor. * - * @param val An enuam value of script type \c Analyzer::Tag. + * @param val An enum value of script type \c Analyzer::Tag. */ - Tag(EnumVal* val); - -private: - type_t type; // Main type. - subtype_t subtype; // Subtype. - mutable EnumVal* val; // Analyzer::Tag value. + Tag(EnumVal* val) : ::Tag(val) {} }; } diff --git a/src/file_analysis/CMakeLists.txt b/src/file_analysis/CMakeLists.txt index f22c293cc4..709790cfaf 100644 --- a/src/file_analysis/CMakeLists.txt +++ b/src/file_analysis/CMakeLists.txt @@ -14,6 +14,7 @@ set(file_analysis_SRCS Analyzer.h AnalyzerSet.cc Component.cc + Tag.cc ) bif_target(file_analysis.bif) diff --git a/src/file_analysis/Component.cc b/src/file_analysis/Component.cc index 99531e40f5..8ddd9cceaf 100644 --- a/src/file_analysis/Component.cc +++ b/src/file_analysis/Component.cc @@ -8,17 +8,17 @@ using namespace file_analysis; -analyzer::Tag::type_t Component::type_counter = 0; +file_analysis::Tag::type_t Component::type_counter = 0; Component::Component(const char* arg_name, factory_callback arg_factory, - analyzer::Tag::subtype_t arg_subtype) + file_analysis::Tag::subtype_t arg_subtype) : plugin::Component(plugin::component::FILE_ANALYZER) { name = copy_string(arg_name); canon_name = canonify_name(arg_name); factory = arg_factory; - tag = analyzer::Tag(++type_counter, arg_subtype); + tag = file_analysis::Tag(++type_counter, arg_subtype); } Component::Component(const Component& other) @@ -36,7 +36,7 @@ Component::~Component() delete [] canon_name; } -analyzer::Tag Component::Tag() const +file_analysis::Tag Component::Tag() const { return tag; } diff --git a/src/file_analysis/Component.h b/src/file_analysis/Component.h index 3cdc69efdf..bd690bc081 100644 --- a/src/file_analysis/Component.h +++ b/src/file_analysis/Component.h @@ -3,7 +3,7 @@ #ifndef FILE_ANALYZER_PLUGIN_COMPONENT_H #define FILE_ANALYZER_PLUGIN_COMPONENT_H -#include "analyzer/Tag.h" +#include "Tag.h" #include "plugin/Component.h" #include "Val.h" @@ -41,12 +41,12 @@ public: * * @param subtype A subtype associated with this component that * further distinguishes it. The subtype will be integrated into - * the analyzer::Tag that the manager associates with this analyzer, - * and analyzer instances can accordingly access it via analyzer::Tag(). - * If not used, leave at zero. + * the file_analysis::Tag that the manager associates with this analyzer, + * and analyzer instances can accordingly access it via + * file_analysis::Tag(). If not used, leave at zero. */ Component(const char* name, factory_callback factory, - analyzer::Tag::subtype_t subtype = 0); + file_analysis::Tag::subtype_t subtype = 0); /** * Copy constructor. @@ -84,7 +84,7 @@ public: * generated for each new Components, and hence unique across all of * them. */ - analyzer::Tag Tag() const; + file_analysis::Tag Tag() const; /** * Generates a human-readable description of the component's main @@ -98,10 +98,10 @@ private: const char* name; // The analyzer's name. const char* canon_name; // The analyzer's canonical name. factory_callback factory; // The analyzer's factory callback. - analyzer::Tag tag; // The automatically assigned analyzer tag. + file_analysis::Tag tag; // The automatically assigned analyzer tag. // Global counter used to generate unique tags. - static analyzer::Tag::type_t type_counter; + static file_analysis::Tag::type_t type_counter; }; } diff --git a/src/file_analysis/Manager.h b/src/file_analysis/Manager.h index a93e78c638..55ff0896d7 100644 --- a/src/file_analysis/Manager.h +++ b/src/file_analysis/Manager.h @@ -294,7 +294,7 @@ protected: private: typedef map analyzer_map_by_name; - typedef map analyzer_map_by_tag; + typedef map analyzer_map_by_tag; typedef map analyzer_map_by_val; void RegisterAnalyzerComponent(Component* component); diff --git a/src/file_analysis/Tag.cc b/src/file_analysis/Tag.cc new file mode 100644 index 0000000000..6f0774a4b4 --- /dev/null +++ b/src/file_analysis/Tag.cc @@ -0,0 +1,24 @@ +// See the file "COPYING" in the main distribution directory for copyright. + +#include "Tag.h" +#include "Manager.h" + +using namespace file_analysis; + +file_analysis::Tag file_analysis::Tag::Error; + +file_analysis::Tag::Tag(type_t type, subtype_t subtype) + : ::Tag(file_mgr->GetTagEnumType(), type, subtype) + { + } + +file_analysis::Tag& file_analysis::Tag::operator=(const file_analysis::Tag& other) + { + ::Tag::operator=(other); + return *this; + } + +EnumVal* file_analysis::Tag::AsEnumVal() const + { + return ::Tag::AsEnumVal(file_mgr->GetTagEnumType()); + } diff --git a/src/file_analysis/Tag.h b/src/file_analysis/Tag.h new file mode 100644 index 0000000000..85c20da5b5 --- /dev/null +++ b/src/file_analysis/Tag.h @@ -0,0 +1,115 @@ +// See the file "COPYING" in the main distribution directory for copyright. + +#ifndef FILE_ANALYZER_TAG_H +#define FILE_ANALYZER_TAG_H + +#include "config.h" +#include "util.h" +#include "../Tag.h" + +class EnumVal; + +namespace file_analysis { + +class Manager; +class Component; + +/** + * Class to identify a file analyzer type. + * + * The script-layer analogue is Files::Tag. + */ +class Tag : public ::Tag { +public: + /* + * Copy constructor. + */ + Tag(const Tag& other) : ::Tag(other) {} + + /** + * Default constructor. This initializes the tag with an error value + * that will make \c operator \c bool return false. + */ + Tag() : ::Tag() {} + + /** + * Destructor. + */ + ~Tag() {} + + /** + * Returns false if the tag represents an error value rather than a + * legal analyzer type. + * TODO: make this conversion operator "explicit" (C++11) or use a + * "safe bool" idiom (not necessary if "explicit" is available), + * otherwise this may allow nonsense/undesired comparison operations. + * + */ + operator bool() const { return *this != Tag(); } + + /** + * Assignment operator. + */ + Tag& operator=(const Tag& other); + + /** + * Compares two tags for equality. + */ + bool operator==(const Tag& other) const + { + return ::Tag::operator==(other); + } + + /** + * Compares two tags for inequality. + */ + bool operator!=(const Tag& other) const + { + return ::Tag::operator!=(other); + } + + /** + * Compares two tags for less-than relationship. + */ + bool operator<(const Tag& other) const + { + return ::Tag::operator<(other); + } + + /** + * Returns the \c Files::Tag enum that corresponds to this tag. + * The returned value does not have its ref-count increased. + * + * @param etype the script-layer enum type associated with the tag. + */ + EnumVal* AsEnumVal() const; + + static Tag Error; + +protected: + friend class file_analysis::Manager; + friend class file_analysis::Component; + + /** + * Constructor. + * + * @param type The main type. Note that the \a file_analysis::Manager + * manages the value space internally, so noone else should assign + * main tyoes. + * + * @param subtype The sub type, which is left to an analyzer for + * interpretation. By default it's set to zero. + */ + Tag(type_t type, subtype_t subtype = 0); + + /** + * Constructor. + * + * @param val An enum value of script type \c Files::Tag. + */ + Tag(EnumVal* val) : ::Tag(val) {} +}; + +} + +#endif From 5fa9c5865b6748c642d91a01f3537331ee5747a9 Mon Sep 17 00:00:00 2001 From: Jon Siwek Date: Wed, 31 Jul 2013 09:48:19 -0500 Subject: [PATCH 3/6] Factor out the need for a tag field in Files::AnalyzerArgs record. This cleans up internals of how analyzer instances get identified by the tag plus any args given to it and doesn't change script code a user would write. --- scripts/base/frameworks/files/main.bro | 12 +--- src/file_analysis/Analyzer.cc | 11 ++++ src/file_analysis/Analyzer.h | 30 ++-------- src/file_analysis/AnalyzerSet.cc | 56 ++++++++++--------- src/file_analysis/AnalyzerSet.h | 26 ++++++--- src/file_analysis/CMakeLists.txt | 2 +- src/file_analysis/File.cc | 18 +++--- src/file_analysis/File.h | 7 ++- src/file_analysis/Manager.cc | 56 ++++++++++++++----- src/file_analysis/Manager.h | 42 ++++++++++++-- .../analyzer/data_event/DataEvent.cc | 5 +- src/file_analysis/analyzer/extract/Extract.cc | 4 +- src/file_analysis/analyzer/hash/Hash.cc | 3 +- src/file_analysis/file_analysis.bif | 12 ++-- 14 files changed, 177 insertions(+), 107 deletions(-) create mode 100644 src/file_analysis/Analyzer.cc diff --git a/scripts/base/frameworks/files/main.bro b/scripts/base/frameworks/files/main.bro index a87608054d..c1883e037f 100644 --- a/scripts/base/frameworks/files/main.bro +++ b/scripts/base/frameworks/files/main.bro @@ -228,11 +228,6 @@ redef record fa_file += { info: Info &optional; }; -redef record AnalyzerArgs += { - # This is used interally for the core file analyzer api. - tag: Files::Tag &optional; -}; - # Store the callbacks for protocol analyzers that have files. global registered_protocols: table[Analyzer::Tag] of ProtoRegistration = table(); @@ -275,14 +270,12 @@ function set_timeout_interval(f: fa_file, t: interval): bool function add_analyzer(f: fa_file, tag: Files::Tag, args: AnalyzerArgs): bool { - # This is to construct the correct args for the core API. - args$tag = tag; add f$info$analyzers[Files::analyzer_name(tag)]; if ( tag in analyzer_add_callbacks ) analyzer_add_callbacks[tag](f, args); - if ( ! __add_analyzer(f$id, args) ) + if ( ! __add_analyzer(f$id, tag, args) ) { Reporter::warning(fmt("Analyzer %s not added successfully to file %s.", tag, f$id)); return F; @@ -297,8 +290,7 @@ function register_analyzer_add_callback(tag: Files::Tag, callback: function(f: f function remove_analyzer(f: fa_file, tag: Files::Tag, args: AnalyzerArgs): bool { - args$tag = tag; - return __remove_analyzer(f$id, args); + return __remove_analyzer(f$id, tag, args); } function stop(f: fa_file): bool diff --git a/src/file_analysis/Analyzer.cc b/src/file_analysis/Analyzer.cc new file mode 100644 index 0000000000..d472f4c80c --- /dev/null +++ b/src/file_analysis/Analyzer.cc @@ -0,0 +1,11 @@ +// See the file "COPYING" in the main distribution directory for copyright. + +#include "Analyzer.h" +#include "Manager.h" + +file_analysis::Analyzer::~Analyzer() + { + DBG_LOG(DBG_FILE_ANALYSIS, "Destroy file analyzer %s", + file_mgr->GetAnalyzerName(tag)); + Unref(args); + } diff --git a/src/file_analysis/Analyzer.h b/src/file_analysis/Analyzer.h index 0a5aa9e25c..e20e2802cf 100644 --- a/src/file_analysis/Analyzer.h +++ b/src/file_analysis/Analyzer.h @@ -5,14 +5,12 @@ #include "Val.h" #include "NetVar.h" -#include "analyzer/Tag.h" +#include "Tag.h" #include "file_analysis/file_analysis.bif.h" namespace file_analysis { -typedef int FA_Tag; - class File; /** @@ -25,11 +23,7 @@ public: * Destructor. Nothing special about it. Virtual since we definitely expect * to delete instances of derived classes via pointers to this class. */ - virtual ~Analyzer() - { - DBG_LOG(DBG_FILE_ANALYSIS, "Destroy file analyzer %d", tag); - Unref(args); - } + virtual ~Analyzer(); /** * Subclasses may override this metod to receive file data non-sequentially. @@ -76,7 +70,7 @@ public: /** * @return the analyzer type enum value. */ - FA_Tag Tag() const { return tag; } + file_analysis::Tag Tag() const { return tag; } /** * @return the AnalyzerArgs associated with the analyzer. @@ -88,18 +82,6 @@ public: */ File* GetFile() const { return file; } - /** - * Retrieves an analyzer tag field from full analyzer argument record. - * @param args an \c AnalyzerArgs (script-layer type) value. - * @return the analyzer tag equivalent of the 'tag' field from the - * \c AnalyzerArgs value \a args. - */ - static FA_Tag ArgsTag(const RecordVal* args) - { - using BifType::Record::Files::AnalyzerArgs; - return args->Lookup(AnalyzerArgs->FieldOffset("tag"))->AsEnum(); - } - protected: /** @@ -108,15 +90,15 @@ protected: * tunable options, if any, related to a particular analyzer type. * @param arg_file the file to which the the analyzer is being attached. */ - Analyzer(RecordVal* arg_args, File* arg_file) - : tag(file_analysis::Analyzer::ArgsTag(arg_args)), + Analyzer(file_analysis::Tag arg_tag, RecordVal* arg_args, File* arg_file) + : tag(arg_tag), args(arg_args->Ref()->AsRecordVal()), file(arg_file) {} private: - FA_Tag tag; /**< The particular analyzer type of the analyzer instance. */ + file_analysis::Tag tag; /**< The particular type of the analyzer instance. */ RecordVal* args; /**< \c AnalyzerArgs val gives tunable analyzer params. */ File* file; /**< The file to which the analyzer is attached. */ }; diff --git a/src/file_analysis/AnalyzerSet.cc b/src/file_analysis/AnalyzerSet.cc index c710d8b085..befb676c87 100644 --- a/src/file_analysis/AnalyzerSet.cc +++ b/src/file_analysis/AnalyzerSet.cc @@ -15,6 +15,7 @@ static void analyzer_del_func(void* v) AnalyzerSet::AnalyzerSet(File* arg_file) : file(arg_file) { TypeList* t = new TypeList(); + t->Append(file_mgr->GetTagEnumType()); t->Append(BifType::Record::Files::AnalyzerArgs->Ref()); analyzer_hash = new CompositeHash(t); Unref(t); @@ -34,20 +35,20 @@ AnalyzerSet::~AnalyzerSet() delete analyzer_hash; } -bool AnalyzerSet::Add(RecordVal* args) +bool AnalyzerSet::Add(file_analysis::Tag tag, RecordVal* args) { - HashKey* key = GetKey(args); + HashKey* key = GetKey(tag, args); if ( analyzer_map.Lookup(key) ) { - DBG_LOG(DBG_FILE_ANALYSIS, "Instantiate analyzer %d skipped for file id" - " %s: already exists", file_analysis::Analyzer::ArgsTag(args), + DBG_LOG(DBG_FILE_ANALYSIS, "Instantiate analyzer %s skipped for file id" + " %s: already exists", file_mgr->GetAnalyzerName(tag), file->GetID().c_str()); delete key; return true; } - file_analysis::Analyzer* a = InstantiateAnalyzer(args); + file_analysis::Analyzer* a = InstantiateAnalyzer(tag, args); if ( ! a ) { @@ -60,10 +61,10 @@ bool AnalyzerSet::Add(RecordVal* args) return true; } -bool AnalyzerSet::QueueAdd(RecordVal* args) +bool AnalyzerSet::QueueAdd(file_analysis::Tag tag, RecordVal* args) { - HashKey* key = GetKey(args); - file_analysis::Analyzer* a = InstantiateAnalyzer(args); + HashKey* key = GetKey(tag, args); + file_analysis::Analyzer* a = InstantiateAnalyzer(tag, args); if ( ! a ) { @@ -80,8 +81,9 @@ bool AnalyzerSet::AddMod::Perform(AnalyzerSet* set) { if ( set->analyzer_map.Lookup(key) ) { - DBG_LOG(DBG_FILE_ANALYSIS, "Add analyzer %d skipped for file id" - " %s: already exists", a->Tag(), a->GetFile()->GetID().c_str()); + DBG_LOG(DBG_FILE_ANALYSIS, "Add analyzer %s skipped for file id" + " %s: already exists", file_mgr->GetAnalyzerName(a->Tag()), + a->GetFile()->GetID().c_str()); Abort(); return true; @@ -91,12 +93,12 @@ bool AnalyzerSet::AddMod::Perform(AnalyzerSet* set) return true; } -bool AnalyzerSet::Remove(const RecordVal* args) +bool AnalyzerSet::Remove(file_analysis::Tag tag, RecordVal* args) { - return Remove(file_analysis::Analyzer::ArgsTag(args), GetKey(args)); + return Remove(tag, GetKey(tag, args)); } -bool AnalyzerSet::Remove(FA_Tag tag, HashKey* key) +bool AnalyzerSet::Remove(file_analysis::Tag tag, HashKey* key) { file_analysis::Analyzer* a = (file_analysis::Analyzer*) analyzer_map.Remove(key); @@ -105,22 +107,22 @@ bool AnalyzerSet::Remove(FA_Tag tag, HashKey* key) if ( ! a ) { - DBG_LOG(DBG_FILE_ANALYSIS, "Skip remove analyzer %d for file id %s", - tag, file->GetID().c_str()); + DBG_LOG(DBG_FILE_ANALYSIS, "Skip remove analyzer %s for file id %s", + file_mgr->GetAnalyzerName(tag), file->GetID().c_str()); return false; } - DBG_LOG(DBG_FILE_ANALYSIS, "Remove analyzer %d for file id %s", a->Tag(), + DBG_LOG(DBG_FILE_ANALYSIS, "Remove analyzer %s for file id %s", + file_mgr->GetAnalyzerName(tag), file->GetID().c_str()); delete a; return true; } -bool AnalyzerSet::QueueRemove(const RecordVal* args) +bool AnalyzerSet::QueueRemove(file_analysis::Tag tag, RecordVal* args) { - HashKey* key = GetKey(args); - FA_Tag tag = file_analysis::Analyzer::ArgsTag(args); + HashKey* key = GetKey(tag, args); mod_queue.push(new RemoveMod(tag, key)); @@ -132,18 +134,22 @@ bool AnalyzerSet::RemoveMod::Perform(AnalyzerSet* set) return set->Remove(tag, key); } -HashKey* AnalyzerSet::GetKey(const RecordVal* args) const +HashKey* AnalyzerSet::GetKey(file_analysis::Tag t, RecordVal* args) const { - HashKey* key = analyzer_hash->ComputeHash(args, 1); + ListVal* lv = new ListVal(TYPE_ANY); + lv->Append(t.AsEnumVal()->Ref()); + lv->Append(args->Ref()); + HashKey* key = analyzer_hash->ComputeHash(lv, 1); + Unref(lv); if ( ! key ) reporter->InternalError("AnalyzerArgs type mismatch"); return key; } -file_analysis::Analyzer* AnalyzerSet::InstantiateAnalyzer(RecordVal* args) const +file_analysis::Analyzer* AnalyzerSet::InstantiateAnalyzer(Tag tag, + RecordVal* args) const { - FA_Tag tag = file_analysis::Analyzer::ArgsTag(args); file_analysis::Analyzer* a = file_mgr->InstantiateAnalyzer(tag, args, file); if ( ! a ) @@ -158,8 +164,8 @@ file_analysis::Analyzer* AnalyzerSet::InstantiateAnalyzer(RecordVal* args) const void AnalyzerSet::Insert(file_analysis::Analyzer* a, HashKey* key) { - DBG_LOG(DBG_FILE_ANALYSIS, "Add analyzer %d for file id %s", a->Tag(), - file->GetID().c_str()); + DBG_LOG(DBG_FILE_ANALYSIS, "Add analyzer %s for file id %s", + file_mgr->GetAnalyzerName(a->Tag()), file->GetID().c_str()); analyzer_map.Insert(key, a); delete key; } diff --git a/src/file_analysis/AnalyzerSet.h b/src/file_analysis/AnalyzerSet.h index 6f14149e30..42a54f4943 100644 --- a/src/file_analysis/AnalyzerSet.h +++ b/src/file_analysis/AnalyzerSet.h @@ -9,6 +9,7 @@ #include "Dict.h" #include "CompHash.h" #include "Val.h" +#include "Tag.h" namespace file_analysis { @@ -38,31 +39,35 @@ public: /** * Attach an analyzer to #file immediately. + * @param tag the analyzer tag of the file analyzer to add. * @param args an \c AnalyzerArgs value which specifies an analyzer. * @return true if analyzer was instantiated/attached, else false. */ - bool Add(RecordVal* args); + bool Add(file_analysis::Tag tag, RecordVal* args); /** * Queue the attachment of an analyzer to #file. + * @param tag the analyzer tag of the file analyzer to add. * @param args an \c AnalyzerArgs value which specifies an analyzer. * @return true if analyzer was able to be instantiated, else false. */ - bool QueueAdd(RecordVal* args); + bool QueueAdd(file_analysis::Tag tag, RecordVal* args); /** * Remove an analyzer from #file immediately. + * @param tag the analyzer tag of the file analyzer to remove. * @param args an \c AnalyzerArgs value which specifies an analyzer. * @return false if analyzer didn't exist and so wasn't removed, else true. */ - bool Remove(const RecordVal* args); + bool Remove(file_analysis::Tag tag, RecordVal* args); /** * Queue the removal of an analyzer from #file. + * @param tag the analyzer tag of the file analyzer to remove. * @param args an \c AnalyzerArgs value which specifies an analyzer. * @return true if analyzer exists at time of call, else false; */ - bool QueueRemove(const RecordVal* args); + bool QueueRemove(file_analysis::Tag tag, RecordVal* args); /** * Perform all queued modifications to the current analyzer set. @@ -91,17 +96,20 @@ protected: /** * Get a hash key which represents an analyzer instance. + * @param tag the file analyzer tag. * @param args an \c AnalyzerArgs value which specifies an analyzer. * @return the hash key calculated from \a args */ - HashKey* GetKey(const RecordVal* args) const; + HashKey* GetKey(file_analysis::Tag tag, RecordVal* args) const; /** * Create an instance of a file analyzer. + * @param tag the tag of a file analyzer. * @param args an \c AnalyzerArgs value which specifies an analyzer. * @return a new file analyzer instance. */ - file_analysis::Analyzer* InstantiateAnalyzer(RecordVal* args) const; + file_analysis::Analyzer* InstantiateAnalyzer(file_analysis::Tag tag, + RecordVal* args) const; /** * Insert an analyzer instance in to the set. @@ -116,7 +124,7 @@ protected: * just used for debugging messages. * @param key the hash key which represents the analyzer's \c AnalyzerArgs. */ - bool Remove(FA_Tag tag, HashKey* key); + bool Remove(file_analysis::Tag tag, HashKey* key); private: @@ -175,14 +183,14 @@ private: * @param arg_a an analyzer instance to add to an analyzer set. * @param arg_key hash key representing the analyzer's \c AnalyzerArgs. */ - RemoveMod(FA_Tag arg_tag, HashKey* arg_key) + RemoveMod(file_analysis::Tag arg_tag, HashKey* arg_key) : Modification(), tag(arg_tag), key(arg_key) {} virtual ~RemoveMod() {} virtual bool Perform(AnalyzerSet* set); virtual void Abort() { delete key; } protected: - FA_Tag tag; + file_analysis::Tag tag; HashKey* key; }; diff --git a/src/file_analysis/CMakeLists.txt b/src/file_analysis/CMakeLists.txt index 709790cfaf..846fc4bf15 100644 --- a/src/file_analysis/CMakeLists.txt +++ b/src/file_analysis/CMakeLists.txt @@ -11,7 +11,7 @@ set(file_analysis_SRCS Manager.cc File.cc FileTimer.cc - Analyzer.h + Analyzer.cc AnalyzerSet.cc Component.cc Tag.cc diff --git a/src/file_analysis/File.cc b/src/file_analysis/File.cc index 7189d90932..3db8ecdbee 100644 --- a/src/file_analysis/File.cc +++ b/src/file_analysis/File.cc @@ -230,14 +230,14 @@ void File::ScheduleInactivityTimer() const timer_mgr->Add(new FileTimer(network_time, id, GetTimeoutInterval())); } -bool File::AddAnalyzer(RecordVal* args) +bool File::AddAnalyzer(file_analysis::Tag tag, RecordVal* args) { - return done ? false : analyzers.QueueAdd(args); + return done ? false : analyzers.QueueAdd(tag, args); } -bool File::RemoveAnalyzer(const RecordVal* args) +bool File::RemoveAnalyzer(file_analysis::Tag tag, RecordVal* args) { - return done ? false : analyzers.QueueRemove(args); + return done ? false : analyzers.QueueRemove(tag, args); } bool File::BufferBOF(const u_char* data, uint64 len) @@ -320,7 +320,7 @@ void File::DataIn(const u_char* data, uint64 len, uint64 offset) while ( (a = analyzers.NextEntry(c)) ) { if ( ! a->DeliverChunk(data, len, offset) ) - analyzers.QueueRemove(a->Args()); + analyzers.QueueRemove(a->Tag(), a->Args()); } analyzers.DrainModifications(); @@ -355,7 +355,7 @@ void File::DataIn(const u_char* data, uint64 len) { if ( ! a->DeliverStream(data, len) ) { - analyzers.QueueRemove(a->Args()); + analyzers.QueueRemove(a->Tag(), a->Args()); continue; } @@ -363,7 +363,7 @@ void File::DataIn(const u_char* data, uint64 len) LookupFieldDefaultCount(missing_bytes_idx); if ( ! a->DeliverChunk(data, len, offset) ) - analyzers.QueueRemove(a->Args()); + analyzers.QueueRemove(a->Tag(), a->Args()); } analyzers.DrainModifications(); @@ -388,7 +388,7 @@ void File::EndOfFile() while ( (a = analyzers.NextEntry(c)) ) { if ( ! a->EndOfFile() ) - analyzers.QueueRemove(a->Args()); + analyzers.QueueRemove(a->Tag(), a->Args()); } FileEvent(file_state_remove); @@ -410,7 +410,7 @@ void File::Gap(uint64 offset, uint64 len) while ( (a = analyzers.NextEntry(c)) ) { if ( ! a->Undelivered(offset, len) ) - analyzers.QueueRemove(a->Args()); + analyzers.QueueRemove(a->Tag(), a->Args()); } if ( FileEventAvailable(file_gap) ) diff --git a/src/file_analysis/File.h b/src/file_analysis/File.h index 794734d24b..12c1e061a8 100644 --- a/src/file_analysis/File.h +++ b/src/file_analysis/File.h @@ -10,6 +10,7 @@ #include "Conn.h" #include "Val.h" +#include "Tag.h" #include "AnalyzerSet.h" #include "BroString.h" @@ -94,17 +95,19 @@ public: /** * Queues attaching an analyzer. Only one analyzer per type can be attached * at a time unless the arguments differ. + * @param tag the analyzer tag of the file analyzer to add. * @param args an \c AnalyzerArgs value representing a file analyzer. * @return false if analyzer can't be instantiated, else true. */ - bool AddAnalyzer(RecordVal* args); + bool AddAnalyzer(file_analysis::Tag tag, RecordVal* args); /** * Queues removal of an analyzer. + * @param tag the analyzer tag of the file analyzer to remove. * @param args an \c AnalyzerArgs value representing a file analyzer. * @return true if analyzer was active at time of call, else false. */ - bool RemoveAnalyzer(const RecordVal* args); + bool RemoveAnalyzer(file_analysis::Tag tag, RecordVal* args); /** * Pass in non-sequential data and deliver to attached analyzers. diff --git a/src/file_analysis/Manager.cc b/src/file_analysis/Manager.cc index fb74a409b4..c7fb1fc3b0 100644 --- a/src/file_analysis/Manager.cc +++ b/src/file_analysis/Manager.cc @@ -206,24 +206,26 @@ bool Manager::SetTimeoutInterval(const string& file_id, double interval) const return true; } -bool Manager::AddAnalyzer(const string& file_id, RecordVal* args) const +bool Manager::AddAnalyzer(const string& file_id, file_analysis::Tag tag, + RecordVal* args) const { File* file = Lookup(file_id); if ( ! file ) return false; - return file->AddAnalyzer(args); + return file->AddAnalyzer(tag, args); } -bool Manager::RemoveAnalyzer(const string& file_id, const RecordVal* args) const +bool Manager::RemoveAnalyzer(const string& file_id, file_analysis::Tag tag, + RecordVal* args) const { File* file = Lookup(file_id); if ( ! file ) return false; - return file->RemoveAnalyzer(args); + return file->RemoveAnalyzer(tag, args); } File* Manager::GetFile(const string& file_id, Connection* conn, @@ -367,13 +369,13 @@ bool Manager::IsDisabled(analyzer::Tag tag) return rval; } -Analyzer* Manager::InstantiateAnalyzer(int tag, RecordVal* args, File* f) const +Analyzer* Manager::InstantiateAnalyzer(Tag tag, RecordVal* args, File* f) const { - analyzer_map_by_val::const_iterator it = analyzers_by_val.find(tag); + analyzer_map_by_tag::const_iterator it = analyzers_by_tag.find(tag); - if ( it == analyzers_by_val.end() ) - reporter->InternalError("cannot instantiate unknown file analyzer: %d", - tag); + if ( it == analyzers_by_tag.end() ) + reporter->InternalError("cannot instantiate unknown file analyzer: %s", + tag.AsString().c_str()); Component* c = it->second; @@ -384,17 +386,43 @@ Analyzer* Manager::InstantiateAnalyzer(int tag, RecordVal* args, File* f) const return c->Factory()(args, f); } -const char* Manager::GetAnalyzerName(int tag) const +const char* Manager::GetAnalyzerName(Val* v) const { - analyzer_map_by_val::const_iterator it = analyzers_by_val.find(tag); + return GetAnalyzerName(file_analysis::Tag(v->AsEnumVal())); + } - if ( it == analyzers_by_val.end() ) - reporter->InternalError("cannot get name of unknown file analyzer: %d", - tag); +const char* Manager::GetAnalyzerName(file_analysis::Tag tag) const + { + analyzer_map_by_tag::const_iterator it = analyzers_by_tag.find(tag); + + if ( it == analyzers_by_tag.end() ) + reporter->InternalError("cannot get name of unknown file analyzer: %s", + tag.AsString().c_str()); return it->second->CanonicalName(); } +file_analysis::Tag Manager::GetAnalyzerTag(const char* name) const + { + analyzer_map_by_name::const_iterator it = analyzers_by_name.find(name); + + if ( it == analyzers_by_name.end() ) + return file_analysis::Tag(); + + return it->second->Tag(); + } + +file_analysis::Tag Manager::GetAnalyzerTag(Val* v) const + { + analyzer_map_by_val::const_iterator it = + analyzers_by_val.find(v->AsEnumVal()->InternalInt()); + + if ( it == analyzers_by_val.end() ) + return file_analysis::Tag(); + + return it->second->Tag(); + } + EnumType* Manager::GetTagEnumType() { return tag_enum_type; diff --git a/src/file_analysis/Manager.h b/src/file_analysis/Manager.h index 55ff0896d7..9a37042669 100644 --- a/src/file_analysis/Manager.h +++ b/src/file_analysis/Manager.h @@ -177,18 +177,22 @@ public: * analyzers of a given type can be attached per file identifier at a time * as long as the arguments differ. * @param file_id the file identifier/hash. + * @param tag the analyzer tag of the file analyzer to add. * @param args a \c AnalyzerArgs value which describes a file analyzer. * @return false if the analyzer failed to be instantiated, else true. */ - bool AddAnalyzer(const string& file_id, RecordVal* args) const; + bool AddAnalyzer(const string& file_id, file_analysis::Tag tag, + RecordVal* args) const; /** * Queue removal of an analyzer for a given file identifier. * @param file_id the file identifier/hash. + * @param tag the analyzer tag of the file analyzer to remove. * @param args a \c AnalyzerArgs value which describes a file analyzer. * @return true if the analyzer is active at the time of call, else false. */ - bool RemoveAnalyzer(const string& file_id, const RecordVal* args) const; + bool RemoveAnalyzer(const string& file_id, file_analysis::Tag tag, + RecordVal* args) const; /** * Tells whether analysis for a file is active or ignored. @@ -204,15 +208,43 @@ public: * @param f The file analzer is to be associated with. * @return The new analyzer instance or null if tag is invalid. */ - Analyzer* InstantiateAnalyzer(int tag, RecordVal* args, File* f) const; + Analyzer* InstantiateAnalyzer(Tag tag, RecordVal* args, File* f) const; /** * Translates a script-level file analyzer tag in to corresponding file * analyzer name. - * @param tag The enum val of a file analyzer. + * @param v The enum val of a file analyzer. * @return The human-readable name of the file analyzer. */ - const char* GetAnalyzerName(int tag) const; + const char* GetAnalyzerName(Val* v) const; + + /** + * Translates a script-level file analyzer tag in to corresponding file + * analyzer name. + * @param tag The analyzer tag of a file analyzer. + * @return The human-readable name of the file analyzer. + */ + const char* GetAnalyzerName(file_analysis::Tag tag) const; + + /** + * Translates an analyzer name into the corresponding tag. + * + * @param name The name. + * + * @return The tag. If the name does not correspond to a valid + * analyzer, the returned tag will evaluate to false. + */ + file_analysis::Tag GetAnalyzerTag(const char* name) const; + + /** + * Translates an analyzer enum value into the corresponding tag. + * + * @param v the enum val of the file analyzer. + * + * @return The tag. If the val does not correspond to a valid + * analyzer, the returned tag will evaluate to false. + */ + file_analysis::Tag GetAnalyzerTag(Val* v) const; /** * Returns the enum type that corresponds to the script-level type diff --git a/src/file_analysis/analyzer/data_event/DataEvent.cc b/src/file_analysis/analyzer/data_event/DataEvent.cc index 1b04111c44..44498f41e1 100644 --- a/src/file_analysis/analyzer/data_event/DataEvent.cc +++ b/src/file_analysis/analyzer/data_event/DataEvent.cc @@ -6,12 +6,15 @@ #include "EventRegistry.h" #include "Event.h" #include "util.h" +#include "file_analysis/Manager.h" using namespace file_analysis; DataEvent::DataEvent(RecordVal* args, File* file, EventHandlerPtr ce, EventHandlerPtr se) - : file_analysis::Analyzer(args, file), chunk_event(ce), stream_event(se) + : file_analysis::Analyzer(file_mgr->GetAnalyzerTag("DATA_EVENT"), + args, file), + chunk_event(ce), stream_event(se) { } diff --git a/src/file_analysis/analyzer/extract/Extract.cc b/src/file_analysis/analyzer/extract/Extract.cc index ef37425003..0de1402939 100644 --- a/src/file_analysis/analyzer/extract/Extract.cc +++ b/src/file_analysis/analyzer/extract/Extract.cc @@ -4,11 +4,13 @@ #include "Extract.h" #include "util.h" +#include "file_analysis/Manager.h" using namespace file_analysis; Extract::Extract(RecordVal* args, File* file, const string& arg_filename) - : file_analysis::Analyzer(args, file), filename(arg_filename) + : file_analysis::Analyzer(file_mgr->GetAnalyzerTag("EXTRACT"), args, file), + filename(arg_filename) { fd = open(filename.c_str(), O_WRONLY | O_CREAT | O_TRUNC, 0666); diff --git a/src/file_analysis/analyzer/hash/Hash.cc b/src/file_analysis/analyzer/hash/Hash.cc index 9835f343b6..12463df8bf 100644 --- a/src/file_analysis/analyzer/hash/Hash.cc +++ b/src/file_analysis/analyzer/hash/Hash.cc @@ -5,11 +5,12 @@ #include "Hash.h" #include "util.h" #include "Event.h" +#include "file_analysis/Manager.h" using namespace file_analysis; Hash::Hash(RecordVal* args, File* file, HashVal* hv, const char* arg_kind) - : file_analysis::Analyzer(args, file), hash(hv), fed(false), kind(arg_kind) + : file_analysis::Analyzer(file_mgr->GetAnalyzerTag(to_upper(string(arg_kind)).c_str()), args, file), hash(hv), fed(false), kind(arg_kind) { hash->Init(); } diff --git a/src/file_analysis/file_analysis.bif b/src/file_analysis/file_analysis.bif index b6c80ac800..7e07ddf6bb 100644 --- a/src/file_analysis/file_analysis.bif +++ b/src/file_analysis/file_analysis.bif @@ -16,21 +16,23 @@ function Files::__set_timeout_interval%(file_id: string, t: interval%): bool %} ## :bro:see:`Files::add_analyzer`. -function Files::__add_analyzer%(file_id: string, args: any%): bool +function Files::__add_analyzer%(file_id: string, tag: Files::Tag, args: any%): bool %{ using BifType::Record::Files::AnalyzerArgs; RecordVal* rv = args->AsRecordVal()->CoerceTo(AnalyzerArgs); - bool result = file_mgr->AddAnalyzer(file_id->CheckString(), rv); + bool result = file_mgr->AddAnalyzer(file_id->CheckString(), + file_mgr->GetAnalyzerTag(tag), rv); Unref(rv); return new Val(result, TYPE_BOOL); %} ## :bro:see:`Files::remove_analyzer`. -function Files::__remove_analyzer%(file_id: string, args: any%): bool +function Files::__remove_analyzer%(file_id: string, tag: Files::Tag, args: any%): bool %{ using BifType::Record::Files::AnalyzerArgs; RecordVal* rv = args->AsRecordVal()->CoerceTo(AnalyzerArgs); - bool result = file_mgr->RemoveAnalyzer(file_id->CheckString(), rv); + bool result = file_mgr->RemoveAnalyzer(file_id->CheckString(), + file_mgr->GetAnalyzerTag(tag) , rv); Unref(rv); return new Val(result, TYPE_BOOL); %} @@ -45,7 +47,7 @@ function Files::__stop%(file_id: string%): bool ## :bro:see:`Files::analyzer_name`. function Files::__analyzer_name%(tag: Files::Tag%) : string %{ - return new StringVal(file_mgr->GetAnalyzerName(tag->InternalInt())); + return new StringVal(file_mgr->GetAnalyzerName(tag)); %} module GLOBAL; From 99c89b42d7b945108fd2d75ad59a89a0b93c2144 Mon Sep 17 00:00:00 2001 From: Jon Siwek Date: Thu, 1 Aug 2013 10:35:47 -0500 Subject: [PATCH 4/6] Internal refactoring of how plugin components are tagged/managed. Made some class templates for code that seemed duplicated between file/protocol tags and managers. Seems like it helps a bit and hopefully can be also be used to transition other things that have enum value "tags" (e.g. logging writers, input readers) to the plugin system. --- src/CMakeLists.txt | 2 + src/DebugLogger.cc | 3 +- src/DebugLogger.h | 1 + src/RuleAction.cc | 10 +- src/analyzer/Analyzer.cc | 10 +- src/analyzer/Component.cc | 19 +- src/analyzer/Component.h | 15 +- src/analyzer/Manager.cc | 114 ++------ src/analyzer/Manager.h | 54 +--- src/analyzer/Tag.h | 5 +- src/analyzer/analyzer.bif | 4 +- src/file_analysis/Analyzer.cc | 2 +- src/file_analysis/AnalyzerSet.cc | 12 +- src/file_analysis/Component.cc | 22 +- src/file_analysis/Component.h | 24 +- src/file_analysis/File.cc | 2 +- src/file_analysis/FileTimer.cc | 2 +- src/file_analysis/Manager.cc | 84 +----- src/file_analysis/Manager.h | 59 +---- src/file_analysis/Tag.h | 9 +- .../analyzer/data_event/DataEvent.cc | 2 +- src/file_analysis/analyzer/extract/Extract.cc | 2 +- src/file_analysis/analyzer/hash/Hash.cc | 2 +- src/file_analysis/file_analysis.bif | 6 +- src/plugin/ComponentManager.h | 248 ++++++++++++++++++ src/plugin/TaggedComponent.h | 85 ++++++ 26 files changed, 432 insertions(+), 366 deletions(-) create mode 100644 src/plugin/ComponentManager.h create mode 100644 src/plugin/TaggedComponent.h diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index 7960579c8a..e64dcbb9f6 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -363,6 +363,8 @@ set(bro_SRCS 3rdparty/sqlite3.c plugin/Component.cc + plugin/ComponentManager.h + plugin/TaggedComponent.h plugin/Manager.cc plugin/Plugin.cc plugin/Macros.h diff --git a/src/DebugLogger.cc b/src/DebugLogger.cc index 380f21aa5f..dc557c4a0a 100644 --- a/src/DebugLogger.cc +++ b/src/DebugLogger.cc @@ -16,7 +16,8 @@ DebugLogger::Stream DebugLogger::streams[NUM_DBGS] = { { "notifiers", 0, false }, { "main-loop", 0, false }, { "dpd", 0, false }, { "tm", 0, false }, { "logging", 0, false }, {"input", 0, false }, - { "threading", 0, false }, { "file_analysis", 0, false } + { "threading", 0, false }, { "file_analysis", 0, false }, + { "plugins", 0, false} }; DebugLogger::DebugLogger(const char* filename) diff --git a/src/DebugLogger.h b/src/DebugLogger.h index e293b326a8..c5744642f5 100644 --- a/src/DebugLogger.h +++ b/src/DebugLogger.h @@ -27,6 +27,7 @@ enum DebugStream { DBG_INPUT, // Input streams DBG_THREADING, // Threading system DBG_FILE_ANALYSIS, // File analysis + DBG_PLUGINS, NUM_DBGS // Has to be last }; diff --git a/src/RuleAction.cc b/src/RuleAction.cc index a13392ee40..ec57c96bd2 100644 --- a/src/RuleAction.cc +++ b/src/RuleAction.cc @@ -40,7 +40,7 @@ RuleActionAnalyzer::RuleActionAnalyzer(const char* arg_analyzer) string str(arg_analyzer); string::size_type pos = str.find(':'); string arg = str.substr(0, pos); - analyzer = analyzer_mgr->GetAnalyzerTag(arg.c_str()); + analyzer = analyzer_mgr->GetComponentTag(arg.c_str()); if ( ! analyzer ) reporter->Warning("unknown analyzer '%s' specified in rule", arg.c_str()); @@ -48,7 +48,7 @@ RuleActionAnalyzer::RuleActionAnalyzer(const char* arg_analyzer) if ( pos != string::npos ) { arg = str.substr(pos + 1); - child_analyzer = analyzer_mgr->GetAnalyzerTag(arg.c_str()); + child_analyzer = analyzer_mgr->GetComponentTag(arg.c_str()); if ( ! child_analyzer ) reporter->Warning("unknown analyzer '%s' specified in rule", arg.c_str()); @@ -60,11 +60,11 @@ RuleActionAnalyzer::RuleActionAnalyzer(const char* arg_analyzer) void RuleActionAnalyzer::PrintDebug() { if ( ! child_analyzer ) - fprintf(stderr, "|%s|\n", analyzer_mgr->GetAnalyzerName(analyzer)); + fprintf(stderr, "|%s|\n", analyzer_mgr->GetComponentName(analyzer)); else fprintf(stderr, "|%s:%s|\n", - analyzer_mgr->GetAnalyzerName(analyzer), - analyzer_mgr->GetAnalyzerName(child_analyzer)); + analyzer_mgr->GetComponentName(analyzer), + analyzer_mgr->GetComponentName(child_analyzer)); } diff --git a/src/analyzer/Analyzer.cc b/src/analyzer/Analyzer.cc index ecd3c9f686..b8b739f3cb 100644 --- a/src/analyzer/Analyzer.cc +++ b/src/analyzer/Analyzer.cc @@ -70,12 +70,12 @@ void AnalyzerTimer::Init(Analyzer* arg_analyzer, analyzer_timer_func arg_timer, Ref(analyzer->Conn()); } -analyzer::ID Analyzer::id_counter = 0;; +analyzer::ID Analyzer::id_counter = 0; const char* Analyzer::GetAnalyzerName() const { assert(tag); - return analyzer_mgr->GetAnalyzerName(tag); + return analyzer_mgr->GetComponentName(tag); } void Analyzer::SetAnalyzerTag(const Tag& arg_tag) @@ -87,7 +87,7 @@ void Analyzer::SetAnalyzerTag(const Tag& arg_tag) bool Analyzer::IsAnalyzer(const char* name) { assert(tag); - return strcmp(analyzer_mgr->GetAnalyzerName(tag), name) == 0; + return strcmp(analyzer_mgr->GetComponentName(tag), name) == 0; } // Used in debugging output. @@ -98,7 +98,7 @@ static string fmt_analyzer(Analyzer* a) Analyzer::Analyzer(const char* name, Connection* conn) { - Tag tag = analyzer_mgr->GetAnalyzerTag(name); + Tag tag = analyzer_mgr->GetComponentTag(name); if ( ! tag ) reporter->InternalError("unknown analyzer name %s; mismatch with tag analyzer::Component?", name); @@ -494,7 +494,7 @@ Analyzer* Analyzer::FindChild(Tag arg_tag) Analyzer* Analyzer::FindChild(const char* name) { - Tag tag = analyzer_mgr->GetAnalyzerTag(name); + Tag tag = analyzer_mgr->GetComponentTag(name); return tag ? FindChild(tag) : 0; } diff --git a/src/analyzer/Component.cc b/src/analyzer/Component.cc index ded0a1a2d5..66ab2213bb 100644 --- a/src/analyzer/Component.cc +++ b/src/analyzer/Component.cc @@ -8,29 +8,26 @@ using namespace analyzer; -analyzer::Tag::type_t Component::type_counter = 0; - Component::Component(const char* arg_name, factory_callback arg_factory, Tag::subtype_t arg_subtype, bool arg_enabled, bool arg_partial) - : plugin::Component(plugin::component::ANALYZER) + : plugin::Component(plugin::component::ANALYZER), + plugin::TaggedComponent(arg_subtype) { name = copy_string(arg_name); canon_name = canonify_name(arg_name); factory = arg_factory; enabled = arg_enabled; partial = arg_partial; - - tag = analyzer::Tag(++type_counter, arg_subtype); } Component::Component(const Component& other) - : plugin::Component(Type()) + : plugin::Component(Type()), + plugin::TaggedComponent(other) { name = copy_string(other.name); canon_name = copy_string(other.canon_name); factory = other.factory; enabled = other.enabled; partial = other.partial; - tag = other.tag; } Component::~Component() @@ -39,11 +36,6 @@ Component::~Component() delete [] canon_name; } -analyzer::Tag Component::Tag() const - { - return tag; - } - void Component::Describe(ODesc* d) const { plugin::Component::Describe(d); @@ -63,13 +55,14 @@ void Component::Describe(ODesc* d) const Component& Component::operator=(const Component& other) { + plugin::TaggedComponent::operator=(other); + if ( &other != this ) { name = copy_string(other.name); factory = other.factory; enabled = other.enabled; partial = other.partial; - tag = other.tag; } return *this; diff --git a/src/analyzer/Component.h b/src/analyzer/Component.h index 9e12ed347e..9bc8b357d7 100644 --- a/src/analyzer/Component.h +++ b/src/analyzer/Component.h @@ -5,6 +5,7 @@ #include "Tag.h" #include "plugin/Component.h" +#include "plugin/TaggedComponent.h" #include "../config.h" #include "../util.h" @@ -21,7 +22,8 @@ class Analyzer; * A plugin can provide a specific protocol analyzer by registering this * analyzer component, describing the analyzer. */ -class Component : public plugin::Component { +class Component : public plugin::Component, + public plugin::TaggedComponent { public: typedef Analyzer* (*factory_callback)(Connection* conn); @@ -100,13 +102,6 @@ public: */ bool Enabled() const { return enabled; } - /** - * Returns the analyzer's tag. Note that this is automatically - * generated for each new Components, and hence unique across all of - * them. - */ - analyzer::Tag Tag() const; - /** * Enables or disables this analyzer. * @@ -128,11 +123,7 @@ private: const char* canon_name; // The analyzer's canonical name. factory_callback factory; // The analyzer's factory callback. bool partial; // True if the analyzer supports partial connections. - analyzer::Tag tag; // The automatically assigned analyzer tag. bool enabled; // True if the analyzer is enabled. - - // Global counter used to generate unique tags. - static analyzer::Tag::type_t type_counter; }; } diff --git a/src/analyzer/Manager.cc b/src/analyzer/Manager.cc index 6eb162f204..82453aef06 100644 --- a/src/analyzer/Manager.cc +++ b/src/analyzer/Manager.cc @@ -60,10 +60,8 @@ bool Manager::ConnIndex::operator<(const ConnIndex& other) const } Manager::Manager() + : plugin::ComponentManager("Analyzer") { - tag_enum_type = new EnumType("Analyzer::Tag"); - ::ID* id = install_ID("Tag", "Analyzer", true, true); - add_type(id, tag_enum_type, 0, 0); } Manager::~Manager() @@ -91,14 +89,14 @@ void Manager::InitPreScript() std::list analyzers = plugin_mgr->Components(); for ( std::list::const_iterator i = analyzers.begin(); i != analyzers.end(); i++ ) - RegisterAnalyzerComponent(*i); + RegisterComponent(*i, "ANALYZER_"); // Cache these tags. - analyzer_backdoor = GetAnalyzerTag("BACKDOOR"); - analyzer_connsize = GetAnalyzerTag("CONNSIZE"); - analyzer_interconn = GetAnalyzerTag("INTERCONN"); - analyzer_stepping = GetAnalyzerTag("STEPPINGSTONE"); - analyzer_tcpstats = GetAnalyzerTag("TCPSTATS"); + analyzer_backdoor = GetComponentTag("BACKDOOR"); + analyzer_connsize = GetComponentTag("CONNSIZE"); + analyzer_interconn = GetComponentTag("INTERCONN"); + analyzer_stepping = GetComponentTag("STEPPINGSTONE"); + analyzer_tcpstats = GetComponentTag("TCPSTATS"); } void Manager::InitPostScript() @@ -109,8 +107,9 @@ void Manager::DumpDebug() { #ifdef DEBUG DBG_LOG(DBG_ANALYZER, "Available analyzers after bro_init():"); - for ( analyzer_map_by_name::const_iterator i = analyzers_by_name.begin(); i != analyzers_by_name.end(); i++ ) - DBG_LOG(DBG_ANALYZER, " %s (%s)", i->second->Name(), IsEnabled(i->second->Tag()) ? "enabled" : "disabled"); + list all_analyzers = GetComponents(); + for ( list::const_iterator i = all_analyzers.begin(); i != all_analyzers.end(); ++i ) + DBG_LOG(DBG_ANALYZER, " %s (%s)", (*i)->Name(), IsEnabled((*i)->Tag()) ? "enabled" : "disabled"); DBG_LOG(DBG_ANALYZER, ""); DBG_LOG(DBG_ANALYZER, "Analyzers by port:"); @@ -120,7 +119,7 @@ void Manager::DumpDebug() string s; for ( tag_set::const_iterator j = i->second->begin(); j != i->second->end(); j++ ) - s += string(GetAnalyzerName(*j)) + " "; + s += string(GetComponentName(*j)) + " "; DBG_LOG(DBG_ANALYZER, " %d/tcp: %s", i->first, s.c_str()); } @@ -130,7 +129,7 @@ void Manager::DumpDebug() string s; for ( tag_set::const_iterator j = i->second->begin(); j != i->second->end(); j++ ) - s += string(GetAnalyzerName(*j)) + " "; + s += string(GetComponentName(*j)) + " "; DBG_LOG(DBG_ANALYZER, " %d/udp: %s", i->first, s.c_str()); } @@ -142,25 +141,6 @@ void Manager::Done() { } -void Manager::RegisterAnalyzerComponent(Component* component) - { - const char* cname = component->CanonicalName(); - - if ( Lookup(cname) ) - reporter->FatalError("Analyzer %s defined more than once", cname); - - DBG_LOG(DBG_ANALYZER, "Registering analyzer %s (tag %s)", - component->Name(), component->Tag().AsString().c_str()); - - analyzers_by_name.insert(std::make_pair(cname, component)); - analyzers_by_tag.insert(std::make_pair(component->Tag(), component)); - analyzers_by_val.insert(std::make_pair(component->Tag().AsEnumVal()->InternalInt(), component)); - - // Install enum "Analyzer::ANALYZER_*" - string id = fmt("ANALYZER_%s", cname); - tag_enum_type->AddName("Analyzer", id.c_str(), component->Tag().AsEnumVal()->InternalInt(), true); - } - bool Manager::EnableAnalyzer(Tag tag) { Component* p = Lookup(tag); @@ -217,8 +197,9 @@ void Manager::DisableAllAnalyzers() { DBG_LOG(DBG_ANALYZER, "Disabling all analyzers"); - for ( analyzer_map_by_tag::const_iterator i = analyzers_by_tag.begin(); i != analyzers_by_tag.end(); i++ ) - i->second->SetEnabled(false); + list all_analyzers = GetComponents(); + for ( list::const_iterator i = all_analyzers.begin(); i != all_analyzers.end(); ++i ) + (*i)->SetEnabled(false); } bool Manager::IsEnabled(Tag tag) @@ -270,7 +251,7 @@ bool Manager::RegisterAnalyzerForPort(Tag tag, TransportProto proto, uint32 port tag_set* l = LookupPort(proto, port, true); #ifdef DEBUG - const char* name = GetAnalyzerName(tag); + const char* name = GetComponentName(tag); DBG_LOG(DBG_ANALYZER, "Registering analyzer %s for port %" PRIu32 "/%d", name, port, proto); #endif @@ -283,7 +264,7 @@ bool Manager::UnregisterAnalyzerForPort(Tag tag, TransportProto proto, uint32 po tag_set* l = LookupPort(proto, port, true); #ifdef DEBUG - const char* name = GetAnalyzerName(tag); + const char* name = GetComponentName(tag); DBG_LOG(DBG_ANALYZER, "Unregistering analyzer %s for port %" PRIu32 "/%d", name, port, proto); #endif @@ -302,7 +283,7 @@ Analyzer* Manager::InstantiateAnalyzer(Tag tag, Connection* conn) return 0; if ( ! c->Factory() ) - reporter->InternalError("analyzer %s cannot be instantiated dynamically", GetAnalyzerName(tag)); + reporter->InternalError("analyzer %s cannot be instantiated dynamically", GetComponentName(tag)); Analyzer* a = c->Factory()(conn); @@ -316,59 +297,10 @@ Analyzer* Manager::InstantiateAnalyzer(Tag tag, Connection* conn) Analyzer* Manager::InstantiateAnalyzer(const char* name, Connection* conn) { - Tag tag = GetAnalyzerTag(name); + Tag tag = GetComponentTag(name); return tag ? InstantiateAnalyzer(tag, conn) : 0; } -const char* Manager::GetAnalyzerName(Tag tag) - { - static const char* error = ""; - - if ( ! tag ) - return error; - - Component* c = Lookup(tag); - - if ( ! c ) - reporter->InternalError("request for name of unknown analyzer tag %s", tag.AsString().c_str()); - - return c->CanonicalName(); - } - -const char* Manager::GetAnalyzerName(Val* val) - { - return GetAnalyzerName(Tag(val->AsEnumVal())); - } - -analyzer::Tag Manager::GetAnalyzerTag(const char* name) - { - Component* c = Lookup(name); - return c ? c->Tag() : Tag(); - } - -EnumType* Manager::GetTagEnumType() - { - return tag_enum_type; - } - -Component* Manager::Lookup(const char* name) - { - analyzer_map_by_name::const_iterator i = analyzers_by_name.find(to_upper(name)); - return i != analyzers_by_name.end() ? i->second : 0; - } - -Component* Manager::Lookup(const Tag& tag) - { - analyzer_map_by_tag::const_iterator i = analyzers_by_tag.find(tag); - return i != analyzers_by_tag.end() ? i->second : 0; - } - -Component* Manager::Lookup(EnumVal* val) - { - analyzer_map_by_val::const_iterator i = analyzers_by_val.find(val->InternalInt()); - return i != analyzers_by_val.end() ? i->second : 0; - } - Manager::tag_set* Manager::LookupPort(TransportProto proto, uint32 port, bool add_if_not_found) { analyzer_map_by_port* m = 0; @@ -461,7 +393,7 @@ bool Manager::BuildInitialAnalyzerTree(Connection* conn) root->AddChildAnalyzer(analyzer, false); DBG_ANALYZER_ARGS(conn, "activated %s analyzer as scheduled", - analyzer_mgr->GetAnalyzerName(*i)); + analyzer_mgr->GetComponentName(*i)); } } @@ -487,7 +419,7 @@ bool Manager::BuildInitialAnalyzerTree(Connection* conn) root->AddChildAnalyzer(analyzer, false); DBG_ANALYZER_ARGS(conn, "activated %s analyzer due to port %d", - analyzer_mgr->GetAnalyzerName(*j), resp_port); + analyzer_mgr->GetComponentName(*j), resp_port); } } } @@ -613,7 +545,7 @@ void Manager::ExpireScheduledAnalyzers() conns.erase(i); DBG_LOG(DBG_ANALYZER, "Expiring expected analyzer %s for connection %s", - analyzer_mgr->GetAnalyzerName(a->analyzer), + analyzer_mgr->GetComponentName(a->analyzer), fmt_conn_id(a->conn.orig, 0, a->conn.resp, a->conn.resp_p)); delete a; @@ -655,7 +587,7 @@ void Manager::ScheduleAnalyzer(const IPAddr& orig, const IPAddr& resp, TransportProto proto, const char* analyzer, double timeout) { - Tag tag = GetAnalyzerTag(analyzer); + Tag tag = GetComponentTag(analyzer); if ( tag != Tag() ) ScheduleAnalyzer(orig, resp, resp_p, proto, tag, timeout); diff --git a/src/analyzer/Manager.h b/src/analyzer/Manager.h index efae629971..d151709eda 100644 --- a/src/analyzer/Manager.h +++ b/src/analyzer/Manager.h @@ -26,6 +26,7 @@ #include "Analyzer.h" #include "Component.h" #include "Tag.h" +#include "plugin/ComponentManager.h" #include "../Dict.h" #include "../net_util.h" @@ -49,7 +50,7 @@ namespace analyzer { * classes. This allows to external analyzer code to potentially use a * different C++ standard library. */ -class Manager { +class Manager : public plugin::ComponentManager { public: /** * Constructor. @@ -231,42 +232,6 @@ public: */ Analyzer* InstantiateAnalyzer(const char* name, Connection* c); - /** - * Translates an analyzer tag into corresponding analyzer name. - * - * @param tag The analyzer tag. - * - * @return The name, or an empty string if the tag is invalid. - */ - const char* GetAnalyzerName(Tag tag); - - /** - * Translates an script-level analyzer tag into corresponding - * analyzer name. - * - * @param val The analyzer tag as an script-level enum value of type - * \c Analyzer::Tag. - * - * @return The name, or an empty string if the tag is invalid. - */ - const char* GetAnalyzerName(Val* val); - - /** - * Translates an analyzer name into the corresponding tag. - * - * @param name The name. - * - * @return The tag. If the name does not correspond to a valid - * analyzer, the returned tag will evaluate to false. - */ - Tag GetAnalyzerTag(const char* name); - - /** - * Returns the enum type that corresponds to the script-level type \c - * Analyzer::Tag. - */ - EnumType* GetTagEnumType(); - /** * Given the first packet of a connection, builds its initial * analyzer tree. @@ -350,18 +315,8 @@ public: private: typedef set tag_set; - typedef map analyzer_map_by_name; - typedef map analyzer_map_by_tag; - typedef map analyzer_map_by_val; typedef map analyzer_map_by_port; - void RegisterAnalyzerComponent(Component* component); // Takes ownership. - - Component* Lookup(const string& name); - Component* Lookup(const char* name); - Component* Lookup(const Tag& tag); - Component* Lookup(EnumVal* val); - tag_set* LookupPort(PortVal* val, bool add_if_not_found); tag_set* LookupPort(TransportProto proto, uint32 port, bool add_if_not_found); @@ -370,9 +325,6 @@ private: analyzer_map_by_port analyzers_by_port_tcp; analyzer_map_by_port analyzers_by_port_udp; - analyzer_map_by_name analyzers_by_name; - analyzer_map_by_tag analyzers_by_tag; - analyzer_map_by_val analyzers_by_val; Tag analyzer_backdoor; Tag analyzer_connsize; @@ -380,8 +332,6 @@ private: Tag analyzer_stepping; Tag analyzer_tcpstats; - EnumType* tag_enum_type; - //// Data structures to track analyzed scheduled for future connections. // The index for a scheduled connection. diff --git a/src/analyzer/Tag.h b/src/analyzer/Tag.h index 8ac151e4b5..d01c8902ee 100644 --- a/src/analyzer/Tag.h +++ b/src/analyzer/Tag.h @@ -6,6 +6,8 @@ #include "config.h" #include "util.h" #include "../Tag.h" +#include "plugin/TaggedComponent.h" +#include "plugin/ComponentManager.h" class EnumVal; @@ -87,7 +89,8 @@ public: protected: friend class analyzer::Manager; - friend class analyzer::Component; + friend class plugin::ComponentManager; + friend class plugin::TaggedComponent; /** * Constructor. diff --git a/src/analyzer/analyzer.bif b/src/analyzer/analyzer.bif index 4d70816075..ebf8083624 100644 --- a/src/analyzer/analyzer.bif +++ b/src/analyzer/analyzer.bif @@ -41,11 +41,11 @@ function Analyzer::__schedule_analyzer%(orig: addr, resp: addr, resp_p: port, function __name%(atype: Analyzer::Tag%) : string %{ - return new StringVal(analyzer_mgr->GetAnalyzerName(atype)); + return new StringVal(analyzer_mgr->GetComponentName(atype)); %} function __tag%(name: string%) : Analyzer::Tag %{ - analyzer::Tag t = analyzer_mgr->GetAnalyzerTag(name->CheckString()); + analyzer::Tag t = analyzer_mgr->GetComponentTag(name->CheckString()); return t.AsEnumVal()->Ref(); %} diff --git a/src/file_analysis/Analyzer.cc b/src/file_analysis/Analyzer.cc index d472f4c80c..e0b5011aa8 100644 --- a/src/file_analysis/Analyzer.cc +++ b/src/file_analysis/Analyzer.cc @@ -6,6 +6,6 @@ file_analysis::Analyzer::~Analyzer() { DBG_LOG(DBG_FILE_ANALYSIS, "Destroy file analyzer %s", - file_mgr->GetAnalyzerName(tag)); + file_mgr->GetComponentName(tag)); Unref(args); } diff --git a/src/file_analysis/AnalyzerSet.cc b/src/file_analysis/AnalyzerSet.cc index befb676c87..2dc4902314 100644 --- a/src/file_analysis/AnalyzerSet.cc +++ b/src/file_analysis/AnalyzerSet.cc @@ -42,7 +42,7 @@ bool AnalyzerSet::Add(file_analysis::Tag tag, RecordVal* args) if ( analyzer_map.Lookup(key) ) { DBG_LOG(DBG_FILE_ANALYSIS, "Instantiate analyzer %s skipped for file id" - " %s: already exists", file_mgr->GetAnalyzerName(tag), + " %s: already exists", file_mgr->GetComponentName(tag), file->GetID().c_str()); delete key; return true; @@ -82,7 +82,7 @@ bool AnalyzerSet::AddMod::Perform(AnalyzerSet* set) if ( set->analyzer_map.Lookup(key) ) { DBG_LOG(DBG_FILE_ANALYSIS, "Add analyzer %s skipped for file id" - " %s: already exists", file_mgr->GetAnalyzerName(a->Tag()), + " %s: already exists", file_mgr->GetComponentName(a->Tag()), a->GetFile()->GetID().c_str()); Abort(); @@ -108,12 +108,12 @@ bool AnalyzerSet::Remove(file_analysis::Tag tag, HashKey* key) if ( ! a ) { DBG_LOG(DBG_FILE_ANALYSIS, "Skip remove analyzer %s for file id %s", - file_mgr->GetAnalyzerName(tag), file->GetID().c_str()); + file_mgr->GetComponentName(tag), file->GetID().c_str()); return false; } DBG_LOG(DBG_FILE_ANALYSIS, "Remove analyzer %s for file id %s", - file_mgr->GetAnalyzerName(tag), + file_mgr->GetComponentName(tag), file->GetID().c_str()); delete a; @@ -155,7 +155,7 @@ file_analysis::Analyzer* AnalyzerSet::InstantiateAnalyzer(Tag tag, if ( ! a ) { reporter->Error("Failed file analyzer %s instantiation for file id %s", - file_mgr->GetAnalyzerName(tag), file->GetID().c_str()); + file_mgr->GetComponentName(tag), file->GetID().c_str()); return 0; } @@ -165,7 +165,7 @@ file_analysis::Analyzer* AnalyzerSet::InstantiateAnalyzer(Tag tag, void AnalyzerSet::Insert(file_analysis::Analyzer* a, HashKey* key) { DBG_LOG(DBG_FILE_ANALYSIS, "Add analyzer %s for file id %s", - file_mgr->GetAnalyzerName(a->Tag()), file->GetID().c_str()); + file_mgr->GetComponentName(a->Tag()), file->GetID().c_str()); analyzer_map.Insert(key, a); delete key; } diff --git a/src/file_analysis/Component.cc b/src/file_analysis/Component.cc index 8ddd9cceaf..9c47f2c75e 100644 --- a/src/file_analysis/Component.cc +++ b/src/file_analysis/Component.cc @@ -8,26 +8,22 @@ using namespace file_analysis; -file_analysis::Tag::type_t Component::type_counter = 0; - -Component::Component(const char* arg_name, factory_callback arg_factory, - file_analysis::Tag::subtype_t arg_subtype) - : plugin::Component(plugin::component::FILE_ANALYZER) +Component::Component(const char* arg_name, factory_callback arg_factory) + : plugin::Component(plugin::component::FILE_ANALYZER), + plugin::TaggedComponent() { name = copy_string(arg_name); canon_name = canonify_name(arg_name); factory = arg_factory; - - tag = file_analysis::Tag(++type_counter, arg_subtype); } Component::Component(const Component& other) - : plugin::Component(Type()) + : plugin::Component(Type()), + plugin::TaggedComponent(other) { name = copy_string(other.name); canon_name = copy_string(other.canon_name); factory = other.factory; - tag = other.tag; } Component::~Component() @@ -36,11 +32,6 @@ Component::~Component() delete [] canon_name; } -file_analysis::Tag Component::Tag() const - { - return tag; - } - void Component::Describe(ODesc* d) const { plugin::Component::Describe(d); @@ -58,11 +49,12 @@ void Component::Describe(ODesc* d) const Component& Component::operator=(const Component& other) { + plugin::TaggedComponent::operator=(other); + if ( &other != this ) { name = copy_string(other.name); factory = other.factory; - tag = other.tag; } return *this; diff --git a/src/file_analysis/Component.h b/src/file_analysis/Component.h index bd690bc081..4cf2dced60 100644 --- a/src/file_analysis/Component.h +++ b/src/file_analysis/Component.h @@ -5,6 +5,7 @@ #include "Tag.h" #include "plugin/Component.h" +#include "plugin/TaggedComponent.h" #include "Val.h" @@ -22,7 +23,8 @@ class Analyzer; * A plugin can provide a specific file analyzer by registering this * analyzer component, describing the analyzer. */ -class Component : public plugin::Component { +class Component : public plugin::Component, + public plugin::TaggedComponent { public: typedef Analyzer* (*factory_callback)(RecordVal* args, File* file); @@ -38,15 +40,8 @@ public: * from file_analysis::Analyzer. This is typically a static \c * Instatiate() method inside the class that just allocates and * returns a new instance. - * - * @param subtype A subtype associated with this component that - * further distinguishes it. The subtype will be integrated into - * the file_analysis::Tag that the manager associates with this analyzer, - * and analyzer instances can accordingly access it via - * file_analysis::Tag(). If not used, leave at zero. */ - Component(const char* name, factory_callback factory, - file_analysis::Tag::subtype_t subtype = 0); + Component(const char* name, factory_callback factory); /** * Copy constructor. @@ -79,13 +74,6 @@ public: */ factory_callback Factory() const { return factory; } - /** - * Returns the analyzer's tag. Note that this is automatically - * generated for each new Components, and hence unique across all of - * them. - */ - file_analysis::Tag Tag() const; - /** * Generates a human-readable description of the component's main * parameters. This goes into the output of \c "bro -NN". @@ -98,10 +86,6 @@ private: const char* name; // The analyzer's name. const char* canon_name; // The analyzer's canonical name. factory_callback factory; // The analyzer's factory callback. - file_analysis::Tag tag; // The automatically assigned analyzer tag. - - // Global counter used to generate unique tags. - static file_analysis::Tag::type_t type_counter; }; } diff --git a/src/file_analysis/File.cc b/src/file_analysis/File.cc index a27070174b..1197cd06f6 100644 --- a/src/file_analysis/File.cc +++ b/src/file_analysis/File.cc @@ -88,7 +88,7 @@ File::File(const string& file_id, Connection* conn, analyzer::Tag tag, if ( conn ) { // add source, connection, is_orig fields - SetSource(analyzer_mgr->GetAnalyzerName(tag)); + SetSource(analyzer_mgr->GetComponentName(tag)); val->Assign(is_orig_idx, new Val(is_orig, TYPE_BOOL)); UpdateConnectionFields(conn, is_orig); } diff --git a/src/file_analysis/FileTimer.cc b/src/file_analysis/FileTimer.cc index 575857fd15..6b1d70f136 100644 --- a/src/file_analysis/FileTimer.cc +++ b/src/file_analysis/FileTimer.cc @@ -14,7 +14,7 @@ FileTimer::FileTimer(double t, const string& id, double interval) void FileTimer::Dispatch(double t, int is_expire) { - File* file = file_mgr->Lookup(file_id); + File* file = file_mgr->LookupFile(file_id); if ( ! file ) return; diff --git a/src/file_analysis/Manager.cc b/src/file_analysis/Manager.cc index 243786b83a..b7f4335717 100644 --- a/src/file_analysis/Manager.cc +++ b/src/file_analysis/Manager.cc @@ -18,10 +18,8 @@ TableVal* Manager::disabled = 0; string Manager::salt; Manager::Manager() + : ComponentManager("Files") { - tag_enum_type = new EnumType("Files::Tag"); - ::ID* id = install_ID("Tag", "Files", true, true); - add_type(id, tag_enum_type, 0, 0); } Manager::~Manager() @@ -35,27 +33,7 @@ void Manager::InitPreScript() for ( std::list::const_iterator i = analyzers.begin(); i != analyzers.end(); ++i ) - RegisterAnalyzerComponent(*i); - } - -void Manager::RegisterAnalyzerComponent(Component* component) - { - const char* cname = component->CanonicalName(); - - if ( tag_enum_type->Lookup("Files", cname) != -1 ) - reporter->FatalError("File Analyzer %s defined more than once", cname); - - DBG_LOG(DBG_FILE_ANALYSIS, "Registering analyzer %s (tag %s)", - component->Name(), component->Tag().AsString().c_str()); - - analyzers_by_name.insert(std::make_pair(cname, component)); - analyzers_by_tag.insert(std::make_pair(component->Tag(), component)); - analyzers_by_val.insert(std::make_pair( - component->Tag().AsEnumVal()->InternalInt(), component)); - - string id = fmt("ANALYZER_%s", cname); - tag_enum_type->AddName("Files", id.c_str(), - component->Tag().AsEnumVal()->InternalInt(), true); + RegisterComponent(*i, "ANALYZER_"); } void Manager::InitPostScript() @@ -193,7 +171,7 @@ void Manager::SetSize(uint64 size, analyzer::Tag tag, Connection* conn, bool Manager::SetTimeoutInterval(const string& file_id, double interval) const { - File* file = Lookup(file_id); + File* file = LookupFile(file_id); if ( ! file ) return false; @@ -208,7 +186,7 @@ bool Manager::SetTimeoutInterval(const string& file_id, double interval) const bool Manager::AddAnalyzer(const string& file_id, file_analysis::Tag tag, RecordVal* args) const { - File* file = Lookup(file_id); + File* file = LookupFile(file_id); if ( ! file ) return false; @@ -219,7 +197,7 @@ bool Manager::AddAnalyzer(const string& file_id, file_analysis::Tag tag, bool Manager::RemoveAnalyzer(const string& file_id, file_analysis::Tag tag, RecordVal* args) const { - File* file = Lookup(file_id); + File* file = LookupFile(file_id); if ( ! file ) return false; @@ -257,7 +235,7 @@ File* Manager::GetFile(const string& file_id, Connection* conn, return rval; } -File* Manager::Lookup(const string& file_id) const +File* Manager::LookupFile(const string& file_id) const { IDMap::const_iterator it = id_map.find(file_id); @@ -269,7 +247,7 @@ File* Manager::Lookup(const string& file_id) const void Manager::Timeout(const string& file_id, bool is_terminating) { - File* file = Lookup(file_id); + File* file = LookupFile(file_id); if ( ! file ) return; @@ -370,59 +348,15 @@ bool Manager::IsDisabled(analyzer::Tag tag) Analyzer* Manager::InstantiateAnalyzer(Tag tag, RecordVal* args, File* f) const { - analyzer_map_by_tag::const_iterator it = analyzers_by_tag.find(tag); + Component* c = Lookup(tag); - if ( it == analyzers_by_tag.end() ) + if ( ! c ) reporter->InternalError("cannot instantiate unknown file analyzer: %s", tag.AsString().c_str()); - Component* c = it->second; - if ( ! c->Factory() ) reporter->InternalError("file analyzer %s cannot be instantiated " "dynamically", c->CanonicalName()); return c->Factory()(args, f); } - -const char* Manager::GetAnalyzerName(Val* v) const - { - return GetAnalyzerName(file_analysis::Tag(v->AsEnumVal())); - } - -const char* Manager::GetAnalyzerName(file_analysis::Tag tag) const - { - analyzer_map_by_tag::const_iterator it = analyzers_by_tag.find(tag); - - if ( it == analyzers_by_tag.end() ) - reporter->InternalError("cannot get name of unknown file analyzer: %s", - tag.AsString().c_str()); - - return it->second->CanonicalName(); - } - -file_analysis::Tag Manager::GetAnalyzerTag(const char* name) const - { - analyzer_map_by_name::const_iterator it = analyzers_by_name.find(name); - - if ( it == analyzers_by_name.end() ) - return file_analysis::Tag(); - - return it->second->Tag(); - } - -file_analysis::Tag Manager::GetAnalyzerTag(Val* v) const - { - analyzer_map_by_val::const_iterator it = - analyzers_by_val.find(v->AsEnumVal()->InternalInt()); - - if ( it == analyzers_by_val.end() ) - return file_analysis::Tag(); - - return it->second->Tag(); - } - -EnumType* Manager::GetTagEnumType() - { - return tag_enum_type; - } diff --git a/src/file_analysis/Manager.h b/src/file_analysis/Manager.h index 9a37042669..dcf33edc99 100644 --- a/src/file_analysis/Manager.h +++ b/src/file_analysis/Manager.h @@ -18,7 +18,8 @@ #include "File.h" #include "FileTimer.h" #include "Component.h" - +#include "Tag.h" +#include "plugin/ComponentManager.h" #include "analyzer/Tag.h" #include "file_analysis/file_analysis.bif.h" @@ -28,7 +29,7 @@ namespace file_analysis { /** * Main entry point for interacting with file analysis. */ -class Manager { +class Manager : public plugin::ComponentManager { public: /** @@ -210,48 +211,6 @@ public: */ Analyzer* InstantiateAnalyzer(Tag tag, RecordVal* args, File* f) const; - /** - * Translates a script-level file analyzer tag in to corresponding file - * analyzer name. - * @param v The enum val of a file analyzer. - * @return The human-readable name of the file analyzer. - */ - const char* GetAnalyzerName(Val* v) const; - - /** - * Translates a script-level file analyzer tag in to corresponding file - * analyzer name. - * @param tag The analyzer tag of a file analyzer. - * @return The human-readable name of the file analyzer. - */ - const char* GetAnalyzerName(file_analysis::Tag tag) const; - - /** - * Translates an analyzer name into the corresponding tag. - * - * @param name The name. - * - * @return The tag. If the name does not correspond to a valid - * analyzer, the returned tag will evaluate to false. - */ - file_analysis::Tag GetAnalyzerTag(const char* name) const; - - /** - * Translates an analyzer enum value into the corresponding tag. - * - * @param v the enum val of the file analyzer. - * - * @return The tag. If the val does not correspond to a valid - * analyzer, the returned tag will evaluate to false. - */ - file_analysis::Tag GetAnalyzerTag(Val* v) const; - - /** - * Returns the enum type that corresponds to the script-level type - * \c Files::Tag. - */ - EnumType* GetTagEnumType(); - protected: friend class FileTimer; @@ -285,7 +244,7 @@ protected: * @return the File object mapped to \a file_id, or a null pointer if no * mapping exists. */ - File* Lookup(const string& file_id) const; + File* LookupFile(const string& file_id) const; /** * Evaluate timeout policy for a file and remove the File object mapped to @@ -325,20 +284,10 @@ protected: static bool IsDisabled(analyzer::Tag tag); private: - typedef map analyzer_map_by_name; - typedef map analyzer_map_by_tag; - typedef map analyzer_map_by_val; - - void RegisterAnalyzerComponent(Component* component); IDMap id_map; /**< Map file ID to file_analysis::File records. */ IDSet ignored; /**< Ignored files. Will be finally removed on EOF. */ string current_file_id; /**< Hash of what get_file_handle event sets. */ - EnumType* tag_enum_type; /**< File analyzer tag type. */ - - analyzer_map_by_name analyzers_by_name; - analyzer_map_by_tag analyzers_by_tag; - analyzer_map_by_val analyzers_by_val; static TableVal* disabled; /**< Table of disabled analyzers. */ static string salt; /**< A salt added to file handles before hashing. */ diff --git a/src/file_analysis/Tag.h b/src/file_analysis/Tag.h index 85c20da5b5..aa38836403 100644 --- a/src/file_analysis/Tag.h +++ b/src/file_analysis/Tag.h @@ -6,12 +6,13 @@ #include "config.h" #include "util.h" #include "../Tag.h" +#include "plugin/TaggedComponent.h" +#include "plugin/ComponentManager.h" class EnumVal; namespace file_analysis { -class Manager; class Component; /** @@ -87,15 +88,15 @@ public: static Tag Error; protected: - friend class file_analysis::Manager; - friend class file_analysis::Component; + friend class plugin::ComponentManager; + friend class plugin::TaggedComponent; /** * Constructor. * * @param type The main type. Note that the \a file_analysis::Manager * manages the value space internally, so noone else should assign - * main tyoes. + * main types. * * @param subtype The sub type, which is left to an analyzer for * interpretation. By default it's set to zero. diff --git a/src/file_analysis/analyzer/data_event/DataEvent.cc b/src/file_analysis/analyzer/data_event/DataEvent.cc index 44498f41e1..cf2d7e52ec 100644 --- a/src/file_analysis/analyzer/data_event/DataEvent.cc +++ b/src/file_analysis/analyzer/data_event/DataEvent.cc @@ -12,7 +12,7 @@ using namespace file_analysis; DataEvent::DataEvent(RecordVal* args, File* file, EventHandlerPtr ce, EventHandlerPtr se) - : file_analysis::Analyzer(file_mgr->GetAnalyzerTag("DATA_EVENT"), + : file_analysis::Analyzer(file_mgr->GetComponentTag("DATA_EVENT"), args, file), chunk_event(ce), stream_event(se) { diff --git a/src/file_analysis/analyzer/extract/Extract.cc b/src/file_analysis/analyzer/extract/Extract.cc index 0de1402939..28b5cf5a63 100644 --- a/src/file_analysis/analyzer/extract/Extract.cc +++ b/src/file_analysis/analyzer/extract/Extract.cc @@ -9,7 +9,7 @@ using namespace file_analysis; Extract::Extract(RecordVal* args, File* file, const string& arg_filename) - : file_analysis::Analyzer(file_mgr->GetAnalyzerTag("EXTRACT"), args, file), + : file_analysis::Analyzer(file_mgr->GetComponentTag("EXTRACT"), args, file), filename(arg_filename) { fd = open(filename.c_str(), O_WRONLY | O_CREAT | O_TRUNC, 0666); diff --git a/src/file_analysis/analyzer/hash/Hash.cc b/src/file_analysis/analyzer/hash/Hash.cc index 12463df8bf..9829934301 100644 --- a/src/file_analysis/analyzer/hash/Hash.cc +++ b/src/file_analysis/analyzer/hash/Hash.cc @@ -10,7 +10,7 @@ using namespace file_analysis; Hash::Hash(RecordVal* args, File* file, HashVal* hv, const char* arg_kind) - : file_analysis::Analyzer(file_mgr->GetAnalyzerTag(to_upper(string(arg_kind)).c_str()), args, file), hash(hv), fed(false), kind(arg_kind) + : file_analysis::Analyzer(file_mgr->GetComponentTag(to_upper(arg_kind).c_str()), args, file), hash(hv), fed(false), kind(arg_kind) { hash->Init(); } diff --git a/src/file_analysis/file_analysis.bif b/src/file_analysis/file_analysis.bif index 7e07ddf6bb..0e904f298f 100644 --- a/src/file_analysis/file_analysis.bif +++ b/src/file_analysis/file_analysis.bif @@ -21,7 +21,7 @@ function Files::__add_analyzer%(file_id: string, tag: Files::Tag, args: any%): b using BifType::Record::Files::AnalyzerArgs; RecordVal* rv = args->AsRecordVal()->CoerceTo(AnalyzerArgs); bool result = file_mgr->AddAnalyzer(file_id->CheckString(), - file_mgr->GetAnalyzerTag(tag), rv); + file_mgr->GetComponentTag(tag), rv); Unref(rv); return new Val(result, TYPE_BOOL); %} @@ -32,7 +32,7 @@ function Files::__remove_analyzer%(file_id: string, tag: Files::Tag, args: any%) using BifType::Record::Files::AnalyzerArgs; RecordVal* rv = args->AsRecordVal()->CoerceTo(AnalyzerArgs); bool result = file_mgr->RemoveAnalyzer(file_id->CheckString(), - file_mgr->GetAnalyzerTag(tag) , rv); + file_mgr->GetComponentTag(tag) , rv); Unref(rv); return new Val(result, TYPE_BOOL); %} @@ -47,7 +47,7 @@ function Files::__stop%(file_id: string%): bool ## :bro:see:`Files::analyzer_name`. function Files::__analyzer_name%(tag: Files::Tag%) : string %{ - return new StringVal(file_mgr->GetAnalyzerName(tag)); + return new StringVal(file_mgr->GetComponentName(tag)); %} module GLOBAL; diff --git a/src/plugin/ComponentManager.h b/src/plugin/ComponentManager.h new file mode 100644 index 0000000000..16f9d80743 --- /dev/null +++ b/src/plugin/ComponentManager.h @@ -0,0 +1,248 @@ +#ifndef PLUGIN_COMPONENT_MANAGER_H +#define PLUGIN_COMPONENT_MANAGER_H + +#include +#include +#include + +#include "Type.h" +#include "ID.h" +#include "Var.h" +#include "Val.h" +#include "Reporter.h" + +namespace plugin { + +/** + * A class that manages tracking of plugin components (e.g. analyzers) and + * installs identifiers in the script-layer to identify them by a unique tag, + * (a script-layer enum value). + * + * @tparam T A ::Tag type or derivative. + * @tparam C A plugin::TaggedComponent type derivative. + */ +template +class ComponentManager { +public: + + /** + * Constructor creates a new enum type called a "Tag" to associate with + * a component. + * + * @param module The script-layer module in which to install the "Tag" ID + * representing an enum type. + */ + ComponentManager(const string& module); + + /** + * @return The script-layer module in which the component's "Tag" ID lives. + */ + const char* GetModule() const; + + /** + * @return A list of all registered components. + */ + list GetComponents() const; + + /** + * @return The enum type associated with the script-layer "Tag". + */ + EnumType* GetTagEnumType() const; + + /** + * Get a component name from its tag. + * + * @param tag A component's tag. + * @return The canonical component name. + */ + const char* GetComponentName(T tag) const; + + /** + * Get a component name from it's enum value. + * + * @param val A component's enum value. + * @return The canonical component name. + */ + const char* GetComponentName(Val* val) const; + + /** + * Get a component tag from its name. + * + * @param name A component's canonical name. + * @return The component's tag, or a tag representing an error if + * no such component assoicated with the name exists. + */ + T GetComponentTag(const string& name) const; + + /** + * Get a component tag from its enum value. + * + * @param v A component's enum value. + * @return The component's tag, or a tag representing an error if + * no such component assoicated with the value exists. + */ + T GetComponentTag(Val* v) const; + +protected: + + /** + * Add a component the internal maps used to keep track of it and create + * a script-layer ID for the component's enum value. + * + * @param component A component to track. + * @param prefix The script-layer ID associated with the component's enum + * value will be a concatenation of this prefix and the component's + * canonical name. + */ + void RegisterComponent(C* component, const string& prefix = ""); + + /** + * @param name The canonical name of a component. + * @return The component associated with the name or a null pointer if no + * such component exists. + */ + C* Lookup(const string& name) const; + + /** + * @param name A component tag. + * @return The component associated with the tag or a null pointer if no + * such component exists. + */ + C* Lookup(const T& tag) const; + + /** + * @param name A component's enum value. + * @return The component associated with the value or a null pointer if no + * such component exists. + */ + C* Lookup(EnumVal* val) const; + +private: + + string module; /**< Script layer module in which component tags live. */ + EnumType* tag_enum_type; /**< Enum type of component tags. */ + map components_by_name; + map components_by_tag; + map components_by_val; +}; + +template +ComponentManager::ComponentManager(const string& arg_module) + : module(arg_module) + { + tag_enum_type = new EnumType(module + "::Tag"); + ::ID* id = install_ID("Tag", module.c_str(), true, true); + add_type(id, tag_enum_type, 0, 0); + } + +template +const char* ComponentManager::GetModule() const + { + return module.c_str(); + } + +template +list ComponentManager::GetComponents() const + { + list rval; + typename map::const_iterator i; + + for ( i = components_by_tag.begin(); i != components_by_tag.end(); ++i ) + rval.push_back(i->second); + + return rval; + } + +template +EnumType* ComponentManager::GetTagEnumType() const + { + return tag_enum_type; + } + +template +const char* ComponentManager::GetComponentName(T tag) const + { + static const char* error = ""; + + if ( ! tag ) + return error; + + C* c = Lookup(tag); + + if ( ! c ) + reporter->InternalError("request for name of unknown component tag %s", + tag.AsString().c_str()); + + return c->CanonicalName(); + } + +template +const char* ComponentManager::GetComponentName(Val* val) const + { + return GetComponentName(T(val->AsEnumVal())); + } + +template +T ComponentManager::GetComponentTag(const string& name) const + { + C* c = Lookup(name); + return c ? c->Tag() : T(); + } + +template +T ComponentManager::GetComponentTag(Val* v) const + { + C* c = Lookup(v->AsEnumVal()); + return c ? c->Tag() : T(); + } + +template +C* ComponentManager::Lookup(const string& name) const + { + typename map::const_iterator i = + components_by_name.find(to_upper(name)); + return i != components_by_name.end() ? i->second : 0; + } + +template +C* ComponentManager::Lookup(const T& tag) const + { + typename map::const_iterator i = components_by_tag.find(tag); + return i != components_by_tag.end() ? i->second : 0; + } + +template +C* ComponentManager::Lookup(EnumVal* val) const + { + typename map::const_iterator i = + components_by_val.find(val->InternalInt()); + return i != components_by_val.end() ? i->second : 0; + } + +template +void ComponentManager::RegisterComponent(C* component, + const string& prefix) + { + const char* cname = component->CanonicalName(); + + if ( Lookup(cname) ) + reporter->FatalError("Component '%s::%s' defined more than once", + module.c_str(), cname); + + DBG_LOG(DBG_PLUGINS, "Registering component %s (tag %s)", + component->Name(), component->Tag().AsString().c_str()); + + components_by_name.insert(std::make_pair(cname, component)); + components_by_tag.insert(std::make_pair(component->Tag(), component)); + components_by_val.insert(std::make_pair( + component->Tag().AsEnumVal()->InternalInt(), component)); + + // Install an identfier for enum value + string id = fmt("%s%s", prefix.c_str(), cname); + tag_enum_type->AddName(module, id.c_str(), + component->Tag().AsEnumVal()->InternalInt(), true); + } + +} // namespace plugin + +#endif diff --git a/src/plugin/TaggedComponent.h b/src/plugin/TaggedComponent.h new file mode 100644 index 0000000000..99eab9f230 --- /dev/null +++ b/src/plugin/TaggedComponent.h @@ -0,0 +1,85 @@ +#ifndef PLUGIN_TAGGED_COMPONENT_H +#define PLUGIN_TAGGED_COMPONENT_H + +namespace plugin { + +/** + * A class which has a tag of a given type associated with it. + * + * @tparam T A ::Tag type or derivative. + */ +template +class TaggedComponent { +public: + + /** + * Constructor creates a unique tag value for this component. + * + * @param subtype A subtype associated with this component that + * further distinguishes it. The subtype will be integrated into + * the Tag that the manager associates with this component, + * and component instances can accordingly access it via Tag(). + * If not used, leave at zero. + */ + TaggedComponent(typename T::subtype_t subtype = 0); + + /** + * Copy constructor. + * + * @param other Another component from which to copy its tag value. + */ + TaggedComponent(const TaggedComponent& other); + + /** + * Assignment operator. + * + * @param other A component to assign. + * @return The assigned object. + */ + TaggedComponent& operator=(const TaggedComponent& other); + + /** + * @return The component's tag. + */ + T Tag() const; + +private: + + T tag; /**< The automatically assigned analyzer tag. */ + static typename T::type_t type_counter; /**< Used to generate globally + unique tags. */ +}; + +template +TaggedComponent::TaggedComponent(typename T::subtype_t subtype) + { + tag = T(++type_counter, subtype); + } + +template +TaggedComponent::TaggedComponent(const TaggedComponent& other) + { + tag = other.tag; + } + +template +TaggedComponent& +TaggedComponent::operator =(const TaggedComponent& other) + { + if ( &other != this ) + tag = other.tag; + + return *this; + } + +template +T TaggedComponent::Tag() const + { + return tag; + } + +template typename T::type_t TaggedComponent::type_counter(0); + +} // namespace plugin + +#endif From 238e4a8d5c24c5cc2c6637902effe4b322a9fae5 Mon Sep 17 00:00:00 2001 From: Jon Siwek Date: Thu, 1 Aug 2013 10:45:44 -0500 Subject: [PATCH 5/6] Minor fix to file/protocol analyzer plugin reference doc. --- src/BroDoc.cc | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/BroDoc.cc b/src/BroDoc.cc index 3cb271bdbf..55dc8ce558 100644 --- a/src/BroDoc.cc +++ b/src/BroDoc.cc @@ -564,7 +564,7 @@ static void WriteAnalyzerTagDefn(FILE* f, EnumType* e, const string& module) dummy_id->MakeType(); list* r = new list(); - r->push_back("Unique identifiers for protocol analyzers."); + r->push_back("Unique identifiers for analyzers."); BroDocObj bdo(dummy_id, r, true); @@ -622,7 +622,7 @@ void CreateFileAnalyzerDoc(const char* filename) FILE* f = fopen(filename, "w"); fprintf(f, "File Analyzer Reference\n"); - fprintf(f, "===========================\n\n"); + fprintf(f, "=======================\n\n"); WriteAnalyzerTagDefn(f, file_mgr->GetTagEnumType(), "Files"); From ee7dba806d4016af59f969237a4a06d1cf158013 Mon Sep 17 00:00:00 2001 From: Jon Siwek Date: Thu, 1 Aug 2013 12:17:51 -0500 Subject: [PATCH 6/6] Fix some build errors. On GCC, some namespace sensitivity and file analyzer plugins now need to link in Analyzer since it's not just a header anymore. --- src/file_analysis/Manager.cc | 3 ++- src/file_analysis/analyzer/data_event/CMakeLists.txt | 2 +- src/file_analysis/analyzer/extract/CMakeLists.txt | 2 +- src/file_analysis/analyzer/hash/CMakeLists.txt | 2 +- 4 files changed, 5 insertions(+), 4 deletions(-) diff --git a/src/file_analysis/Manager.cc b/src/file_analysis/Manager.cc index b7f4335717..5975133356 100644 --- a/src/file_analysis/Manager.cc +++ b/src/file_analysis/Manager.cc @@ -18,7 +18,8 @@ TableVal* Manager::disabled = 0; string Manager::salt; Manager::Manager() - : ComponentManager("Files") + : plugin::ComponentManager("Files") { } diff --git a/src/file_analysis/analyzer/data_event/CMakeLists.txt b/src/file_analysis/analyzer/data_event/CMakeLists.txt index 81551feda2..49e23d49a0 100644 --- a/src/file_analysis/analyzer/data_event/CMakeLists.txt +++ b/src/file_analysis/analyzer/data_event/CMakeLists.txt @@ -4,5 +4,5 @@ include_directories(BEFORE ${CMAKE_CURRENT_SOURCE_DIR} ${CMAKE_CURRENT_BINARY_DIR}) bro_plugin_begin(Bro FileDataEvent) -bro_plugin_cc(DataEvent.cc Plugin.cc) +bro_plugin_cc(DataEvent.cc Plugin.cc ../../Analyzer.cc) bro_plugin_end() diff --git a/src/file_analysis/analyzer/extract/CMakeLists.txt b/src/file_analysis/analyzer/extract/CMakeLists.txt index df3fa2646d..e413196db2 100644 --- a/src/file_analysis/analyzer/extract/CMakeLists.txt +++ b/src/file_analysis/analyzer/extract/CMakeLists.txt @@ -4,5 +4,5 @@ include_directories(BEFORE ${CMAKE_CURRENT_SOURCE_DIR} ${CMAKE_CURRENT_BINARY_DIR}) bro_plugin_begin(Bro FileExtract) -bro_plugin_cc(Extract.cc Plugin.cc) +bro_plugin_cc(Extract.cc Plugin.cc ../../Analyzer.cc) bro_plugin_end() diff --git a/src/file_analysis/analyzer/hash/CMakeLists.txt b/src/file_analysis/analyzer/hash/CMakeLists.txt index 5734740198..0e3143ee05 100644 --- a/src/file_analysis/analyzer/hash/CMakeLists.txt +++ b/src/file_analysis/analyzer/hash/CMakeLists.txt @@ -4,6 +4,6 @@ include_directories(BEFORE ${CMAKE_CURRENT_SOURCE_DIR} ${CMAKE_CURRENT_BINARY_DIR}) bro_plugin_begin(Bro FileHash) -bro_plugin_cc(Hash.cc Plugin.cc) +bro_plugin_cc(Hash.cc Plugin.cc ../../Analyzer.cc) bro_plugin_bif(events.bif) bro_plugin_end()