I fully ascribe to Hudak-style embedded DSL, which really just are well-designed APIs. External DSLs on the other hand are like puppies, they all start out cute and happy, but without exception turn into vicious beasts as they grow up1 (make, XSLT, regular expressions, …).
external DSLs are ugly: I think they are, but so are death and taxes
usually we got log like following: [sys1] ----> [sys2] 00001 MSG_0 0x00010002 0x00030004 [sys2] ----> [sys3] 00002 MSG_1 0x00010002 0x00030004 [sys3] ----> [sys1] 00003 MSG_2 0x00010002 0x00030004 but a diagram like following is more readable: sys1 sys2 sys3 |----->| | MSG_0 [0x00010002 0x00030004] | |----->| MSG_1 [0x00010002 0x00030004] |<------------| MSG_2 [0x00010002 0x00030004] I implemented a simple app to draw this diagram: https://github.com/whunmr/msgflow more complicated diagram generated from log: ss3 kk3 zz1 qq1 qq2 kk1 bb1 aa1 ss1 | | | | | | | | * MSG_0 [1] [0x00010002 0x00030004] | | | | | | | |<----| MSG_1 [2] [0x00010002 0x00030004] | | | | | | | |---->| MSG_2 [3] [0x00010002 0x00030004] | | | | | |<----| | | MSG_3 [4] [0x00010002 0x00030004] | | | | | |---------------->| MSG_4 [5] [0x00010002 0x00030004] | | | | | |<----------------| MSG_5 [6] [0x00010002 0x00030004] | | | | | | | |<----| MSG_6 [7] [0x00010002 0x00030004] | | | | |<----------------------| MSG_7 [8] [0x00010002 0x00030004] | | |<----| | | | | | MSG_8 [9] [0x00010002 0x00030004] | | |---------------------------->| | MSG_9 [10] [0x00010002 0x00030004] | | | | | | | |---->| MSG_2 [11] [0x00010002 0x00030004] | |<----------------------------| | | MSG_3 [12] [0x00010002 0x00030004] | | | | | |---------------->| MSG_4 [13] [0x00010002 0x00030004] | |<----------------------------------------| MSG_5 [14] [0x00010002 0x00030004] |---------------------------------------->| | MSG_6 [15] [0x00010002 0x00030004] |---------------->| | | | | | MSG_7 [16] [0x00010002 0x00030004] | | |<----------------------------------| MSG_8 [17] [0x00010002 0x00030004] | | |---------------------------------->| MSG_9 [18] [0x00010002 0x00030004]
def_mi(Basic_mi_x) {
step(sync_call_1);
step(sync_call_2);
sysd<----------------------------async_step(async_call_1) {
sysd-------------------->ack(msg_ack_1, on_msg_ack_1);
sysd-------------------->ack(msg_ack_2, on_msg_ack_2);
sysd-------------------->ack_ex(msg_ack_4)
{
step(sync_call_nested_1);
sysx<------------async_step(nested_async_call_1)
{
sysx---->ack(nested_msg_ack_1, on_nested_msg_ack_1);
sysx---->ack(nested_msg_ack_2, on_nested_msg_ack_2);
sysx---->ack(msg_ack_1, on_nested_msg_ack_3);
} async_step_end;
} on_msg_ack_ex_end;
}async_step_end;
} def_mi_end;
In following scenario outline, examples will be generated by `generate_example_data_func`:
Feature: test feature Scenario Outline: testing * test_tool __________________* "<foo>" * test_tool *__________________ "<bar>" Examples: | foo | bar | | `generate_example_data_func` | |
implementation of `generate_example_data_func`, and steps:
Given(/^test_tool __________________\* "(.*?)"$/) do |arg1| end Given(/^test_tool \*__________________ "(.*?)"$/) do |arg1| end def generate_example_data_func table table << ["xxx", "yyy"] table << ["aaa", "bbb"] table end
running result:
$ cucumber –expand
Feature: test feature Scenario Outline: testing # features/reading_report/test.feature:2 * test_tool __________________* "<foo>" # features/step_definitions/new_step.rb:1 * test_tool *__________________ "<bar>" # features/step_definitions/new_step.rb:4 Examples: Scenario: | xxx | yyy | # features/reading_report/test.feature:-1 * test_tool __________________* "xxx" # features/step_definitions/new_step.rb:1 * test_tool *__________________ "yyy" # features/step_definitions/new_step.rb:4 Scenario: | aaa | bbb | # features/reading_report/test.feature:-1 * test_tool __________________* "aaa" # features/step_definitions/new_step.rb:1 * test_tool *__________________ "bbb" # features/step_definitions/new_step.rb:4 2 scenarios (2 passed) 4 steps (4 passed) 0m0.028s
To support dynamic generate example data, we need change the scenario_outline.rb`:
def create_examples_table(example_section_and_gherkin_examples) example_section = example_section_and_gherkin_examples[0] gherkin_examples = example_section_and_gherkin_examples[1] examples_location = example_section[0] examples_comment = example_section[1] examples_keyword = example_section[2] examples_title = example_section[3] examples_description = example_section[4] examples_matrix = example_section[5] #CHANGE BEGIN##################################################### #~/.rvm/gems/ruby-1.9.3-p392/gems/cucumber-1.3.14/lib/cucumber/ast/scenario_outline.rb if examples_matrix.length > 1 && examples_matrix[1].length > 1 using_function_to_generate_examples = !(examples_matrix[1][0] =~ /^`.*`$/).nil? if using_function_to_generate_examples function_str = examples_matrix[1][0].gsub!(/`(.*)`/, '\1') + "([" + examples_matrix[0].to_s + "])" examples_matrix = eval(function_str) end end #CHANGE END##################################################### examples_table = OutlineTable.new(examples_matrix, self) ex = Examples.new(examples_location, examples_comment, examples_keyword, examples_title, examples_description, examples_table) ex.gherkin_statement(gherkin_examples) ex end
Make message interaction more readable, by using c++ local class:::: c++ local class: http://publib.boulder.ibm.com/infocenter/lnxpcomp/v8v101/index.jsp?topic=%2Fcom.ibm.xlcpp8l.doc%2Flanguage%2Fref%2Fcplr062.htm ///////////////////////////////////////////////////////////////////////////// #include <iostream> #include <map> using namespace std; #define MESSAGE_SEND 3 #define MESSAGE_TIME_OUT 4 #define MESSAGE_RESPONSE 5 ///////////////////////////////////////////////////////////////////////////// struct Worker { virtual void operator()() = 0; }; ///////////////////////////////////////////////////////////////////////////// struct HandlerPool { void addMsg(int msg, Worker& a) { maps[msg] = &a; } void run(int msg) { if (maps.find(msg) != maps.end()) { (*maps[msg])(); } else { cout << "no handler" << endl; } } map<int, Worker*> maps; }; HandlerPool g_handlePool; ///////////////////////////////////////////////////////////////////////////// #define if_on_msg(msg) IF_ON_MSG(onMsg_##msg) #define IF_ON_MSG(msg_handler_class) \ static struct msg_handler_class : Worker { \ void operator()() #define end_if_on_msg(msg) END_IF_ON_MSG(msg, onMsg_##msg##_handler) #define END_IF_ON_MSG(msg, handler_instance_name) \ } handler_instance_name; \ g_handlePool.addMsg(msg, handler_instance_name) ///////////////////////////////////////////////////////////////////////////// struct Worker_Before_refactor : Worker { void processMsg(/*const Message& msg*/) { //`people can not see which message came first, which is the last one` //`so can not know the message_interaction by reading code` switch (msg) { case MESSAGE_TIME_OUT: .... case MESSAGE_RESPONSE: .... case MESSAGE_XXXX: .... case MESSAGE_YYYY: .... } } }; ///////////////////////////////////////////////////////////////////////////// struct WorkerAfterRefactor : Worker { void operator()() { sender------->receiver(MessageBody); if_on_msg(sender<------------receiver(MESSAGE_TIME_OUT)) { sender------->receiver(MessageBody); if_on_msg(sender<------------receiver(MESSAGE_TIME_OUT)) { cout << "retry still got timeout..." << endl; } end_if_on_msg(MESSAGE_TIME_OUT); } end_if_on_msg(MESSAGE_TIME_OUT); if_on_msg(sender<-----receiver(MESSAGE_RESPONSE)) { cout << "succeeded got ack" << endl; } end_if_on_msg(MESSAGE_RESPONSE); } }; ///////////////////////////////////////////////////////////////////////////// int main(int argc, const char * argv[]) { WorkerAfterRefactor a; g_handlePool.addMsg(MESSAGE_SEND, a); g_handlePool.run(MESSAGE_SEND); g_handlePool.run(MESSAGE_TIME_OUT); g_handlePool.run(MESSAGE_TIME_OUT); g_handlePool.run(MESSAGE_RESPONSE); return 0; } ///////////////////////////////////////////////////////////////////////////// //output: //------>send sca config to slave board //<----timeout... //----->retry send sca config to slave board //<-----retry still got timeout... //<----succeeded got ack // //Process finished with exit code 0
* usually we need test message communication between sub-systems, so the ft maybe looks like following: before refactor::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: struct HeartBeatAckTest : testing::Test { protected: SubSystemA aaa; SubSystemB bbb; Request request; }; TEST_F(HeartBeatAckTest, should_able_to_reply_heart_beat_ack) { MOCKER(shouldReceiveMsg) //using mockcpp .expects(exactly(1)) .with( eq(MSG_HEART_BEAT_ACK) , any() , eq(bbb.address()) , eq(aaa.address())); Msg::send(aaa, bbb, MSG_HEART_BEAT, &request); } //By using operator overload we can make send and receive more easy to read //we can set MOCKER by operator<, --, () and send msg by operator>, --, () after refactor::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: #define expects_msg if(true) TEST_F(AckTest, aaa_should_able_to_reply_heart_beat_ack) { expects_msg { aaa<-----bbb(MSG_HEART_BEAT_ACK); } aaa---->bbb(MSG_HEART_BEAT, &request); } TEST_F(AckTest, bbb_should_reply_heart_beat_ack___AND___forward_heart_beat_to_ccc) { expects_msg { aaa<-----bbb(MSG_HEART_BEAT_ACK); ccc<-----bbb(MSG_HEART_BEAT); } aaa---->bbb(MSG_HEART_BEAT, &request); } more thoughts:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: // the <----- is used to set MOCKER before ----->, can we? put the <---- after -----> , // to make the message interaction more easy to comprehensive. // similar to `RAII`, actually we can defer the send message operation of -----> to the end of scope expects_msg. struct SendAllBufferedMessageAtEndOfScope { SendAllBufferedMessageAtEndOfScope(int i) { } //make it can init from dummy int ~SendAllBufferedMessageAtEndOfScope() { send_all_buffered_message_in_sequence(); } operator bool() { return true; } }; #define expects_msg_interaction if (SendAllBufferedMessageAtEndOfScope sendMsgAtEnd = 1) TEST_F(AckTest, should_able_to_reply_heart_beat_ack) { expects_msg_interaction { aaa ----> bbb (MSG_HEART_BEAT, &request); //buffer the messge aaa <---- bbb (MSG_HEART_BEAT_ACK); //set the MOCKER specification //invisible destructor code of sendMsgAtEnd } } TEST_F(AckTest, bbb_should_reply_heart_beat_ack___AND___forward_heart_beat_to_ccc) { expects_msg_interaction { aaa ----> bbb (MSG_HEART_BEAT, &request); aaa <---- bbb (MSG_HEART_BEAT_ACK); ccc <---- bbb (MSG_HEART_BEAT); } } example code:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: struct SubSystemB { SubSystemB& operator--(int) { return *this; } template<typename T> bool operator >(const T& target) { target.receiveMsg(); return true; } }; struct SubSystemA { const SubSystemA& operator()(int msgId, int* p) { msgId_ = msgId; return *this; } void receiveMsg() const { std::cout << "cc got message: " << msgId_ << std::endl; } int msgId_; }; int main(int argc, const char * argv[]) { SubSystemB bbb; SubSystemA aaa; int param = 0; bbb---->aaa(1, ¶m); return 0; } //another example: TEST_F(AckTest, bbb_should_reply_heart_beat_ack___AND___forward_heart_beat_to_ccc) { expects_msg_interaction( op1 ----------------> xx (MSG_HEART_BEAT, &request); op1 <---------------- xx (MSG_HEART_BEAT_ACK); op2 ------------> xx (MSG_HEART_BEAT, &request); op3 --------> xx (MSG_HEART_BEAT, &request); op3 <-------- xx (MSG_HEART_BEAT_ACK); op4 ----> xx (MSG_HEART_BEAT, &request); op4 <---- xx (MSG_HEART_BEAT_ACK); op1 ----------------> xx (MSG_HEART_BEAT, &request); op1 <---------------- xx (MSG_HEART_BEAT_ACK); op2 ------------> xx (MSG_HEART_BEAT, &request); op3 --------> xx (MSG_HEART_BEAT, &request); op3 <-------- xx (MSG_HEART_BEAT_ACK); op4 ----> xx (MSG_HEART_BEAT, &request); op4 <---- xx (MSG_HEART_BEAT_ACK); ); }
We always need some temporary files to do editing work. for example, we got 10000+ line logs to trace bug in it, first thing to do is create a tmp file. 1. To create a tmp file we'd like to - quick open the files in a directory without naming collision, with short-key `C-; t` - store the tmp files for future search/grep ;;code (defun create-and-open-tmp-file () (interactive) (find-file (s-concat "~/Documents/hongmin.wang/Google Drive/notes/tmp" (format-time-string "%Y%m%d___%H_%M_%S") (uuid-string) ".temp"))) (global-set-key (kbd "C-; t") 'create-and-open-tmp-file) 2. To manage versions of tmp file, we like to put it in a git direcotry and `C-x v i` to register the tmp file to git and `C-x v v` to commit new version of the tmp file to git. 3. sometime, you think the tmp file is no longer an tmp file, do a "Save As..." by `C-x C-w`
* Can not setup env in home. so i can not debug my application at home remotely I am debugging and web application, but I need an ESP fast server to always return XML data for my search, then I can debug things. but i can not access the fast server at home, so i can not work remotely at home. I want have an application to monitor my web application's interactions with fast server, and recording the interacting data(including delays), and can replay/mock the services at home. * python -m SimpleHTTPServer 9000 * http://localhost:9000/null_main.c * sudo justniffer -i lo0 -r -p "port 9000" GET /null_main.c HTTP/1.1 Host: localhost:9000 Connection: keep-alive Cache-Control: max-age=0 Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/27.0.1453.110 Safari/537.36 Accept-Encoding: gzip,deflate,sdch Accept-Language: en-US,en;q=0.8 If-Modified-Since: Sat, 20 Oct 2012 09:17:23 GMT HTTP/1.0 200 OK Server: SimpleHTTP/0.6 Python/2.7.1 Date: Thu, 06 Jun 2013 13:27:41 GMT Content-type: text/plain Content-Length: 46 Last-Modified: Sat, 20 Oct 2012 09:17:23 GMT /* null_main.c */ int main() { return 0; } * TODO need running the mock server according the recorded http data
* debugging tips strace:: http://www.thegeekstuff.com/2011/11/strace-examples/ where is the log files of specified process (e.g. nginx)? @firstly, find your process's pid ubuntu@ip:~ $ ps -ef | grep nginx root 13622 1 0 Mar03 ? 00:00:00 nginx: master process /usr/sbin/nginx www-data 13623 13622 0 Mar03 ? 00:03:19 nginx: worker process www-data 13624 13622 0 Mar03 ? 00:00:00 nginx: worker process www-data 13625 13622 0 Mar03 ? 00:03:19 nginx: worker process www-data 13626 13622 0 Mar03 ? 00:03:19 nginx: worker process ubuntu 19058 18961 0 10:31 pts/0 00:00:00 grep --color=auto nginx @secondly, using lsof to find all handle of that "pid", then grep it ubuntu@ip:~ $ sudo lsof -f -p 13622 | grep log nginx 13622 root 2w REG 202,1 0 145222 /var/log/nginx/error.log nginx 13622 root 5w REG 202,1 0 135398 /var/log/nginx/access.log nginx 13622 root 6w REG 202,1 0 145222 /var/log/nginx/error.log which process is listen on port 80? ubuntu@ip:~ $ sudo lsof -i:80 COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME nginx 13622 root 7u IPv4 27232 0t0 TCP *:http (LISTEN) nginx 13623 www-data 7u IPv4 27232 0t0 TCP *:http (LISTEN) nginx 13624 www-data 7u IPv4 27232 0t0 TCP *:http (LISTEN) nginx 13625 www-data 7u IPv4 27232 0t0 TCP *:http (LISTEN) nginx 13626 www-data 7u IPv4 27232 0t0 TCP *:http (LISTEN) what's my process's environemnt variables? [vagrant@vagrant-centos-6-64 ~] $ ps -ef | grep apache apache 2136 1382 0 10:11 ? 00:00:00 /usr/sbin/httpd [vagrant@vagrant-centos-6-64 ~] $ sudo cat /proc/1382/environ TERM=linuxPATH=/sbin:/usr/sbin:/bin:/usr/binrunlevel=3RUNLEVEL=3LANGSH_SOURCED=1PWD=/LANG=Cprevious=NPREVLEVEL=NCONSOLETYPE=vtSHLVL=3UPSTART_INSTANCE=UPSTART_EVENTS=runlevelUPSTART_JOB=rc_=/usr/sbin/httpd let mysql log the queries? @execute following command in mysql console SET GLOBAL log_output = 'FILE'; Set GLOBAL general_log_file = '/tmp/mysql.log'; SET GLOBAL general_log = 'ON'; @ or directly in bash $ mysql -uroot -pYOUR-PASSWORD -e "SET GLOBAL log_output = 'FILE'; Set GLOBAL general_log_file = '/tmp/mysql.log'; SET GLOBAL general_log = 'ON';" @ then tail /tmp/mysql.log, see the quries against mysql $ tail -f /tmp/mysql.log how to monitor http request/response of this machine? @install justniffer on mac $ brew install boost $ brew install justniffer @ Capture all tcp traffic $ sudo justniffer -i eth0 -r GET /null_main.c HTTP/1.1 Host: localhost:9000 Connection: keep-alive Cache-Control: max-age=0 Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/27.0.1453.110 Safari/537.36 Accept-Encoding: gzip,deflate,sdch Accept-Language: en-US,en;q=0.8 If-Modified-Since: Sat, 20 Oct 2012 09:17:23 GMT HTTP/1.0 200 OK Server: SimpleHTTP/0.6 Python/2.7.1 Date: Thu, 06 Jun 2013 13:27:41 GMT Content-type: text/plain Content-Length: 46 Last-Modified: Sat, 20 Oct 2012 09:17:23 GMT /* null_main.c */ int main() { return 0; } @ Capture only http traffic $ sudo justniffer -i eth0 -r -p "port 80 or port 8080" @Capture smtp traffic (usually using tcp port 25) $ sudo justniffer -i eth0 -r -p "port 25" 220 plecno.com ESMTP Postfix (Ubuntu) EHLO unknown.localnet 250-plecno.com 250-PIPELINING 250-SIZE 250-VRFY 250-ETRN 250-STARTTLS 250-ENHANCEDSTATUSCODES 250-8BITMIME 250 DSN ... --Boundary-00=_ZI47J3FTNXn+25g Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: 7bit <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN" "http://www.w3.org/TR/REC-html40/strict.dtd"> <html> <head> ... How to monitor changes in directory? @run following command, and try to touch/rm files. the results of `watch` will changing. $ watch -d -n 0.1 "ls" How to start a HTTP server on port 3000 for files in directory ~? $ cd ~ $ python -m SimpleHTTPServer 3000 TO BE CONTINUE::::
.JPG)
gradle clean ideagradle jar