Then, we proceed to parse user input on our own.<\/p>\n
Our methods must return true or false \u2013 true for keeping the console open, false for closing it.<\/p>\n
Ideally, for simple tasks, we could work on the engine directly from there, e.g.:<\/p>\n
DCmd_Register( .... Cmd_SkipInstruction));\r\n \r\nCmd_SkipInstruction (...) {\r\n \/\/ usage: skip threadnumber\r\n _script[atoi(argv[1]))->_iP++;\r\n return true;\r\n}<\/pre>\nThis could get messy fast, though.
\nWhat we have going on here, though, is a MVA sort of pattern, which is more or less like its more famous brother MVC, except that the +** is bidirectional \u2013 the adapter can fiddle with the model and vice versa \u2013 the reason is decoupling and ease of mantainance: http:\/\/en.wikipedia.org\/wiki\/Model%E2%80%93view%E2%80%93adapter<\/a><\/p>\nMVC:<\/strong><\/p>\n![\"\"](\"https:\/\/blogs.scummvm.org\/t0by\/wp-content\/uploads\/sites\/43\/2013\/08\/mvc.png\")
<\/a>MVA:<\/strong><\/p>\n![\"\"](\"https:\/\/blogs.scummvm.org\/t0by\/wp-content\/uploads\/sites\/43\/2013\/08\/mva.png\")
<\/a>(Image: https:\/\/www.palantir.com<\/a>)<\/p>\nWe have a middle layer called the DebuggerAdapter that works on the script engine for us, while we need to knownothing about the script engine itself in the console proper.<\/p>\n
The header looks kinda like this:<\/p>\n
namespace Wintermute {\r\nclass DebuggerAdapter {\r\npublic:\r\n DebuggerAdapter(WintermuteEngine *vm);\r\n \/\/ Called by Script (=~Model)\r\n bool triggerBreakpoint(ScScript *script);\r\n bool triggerStep(ScScript *script);\r\n bool triggerWatch(ScScript *script, const char *symbol);\r\n \/\/ Called by Console (~=View)\r\n int addWatch(const char *filename, const char *symbol);\r\n int addBreakpoint(const char *filename, int line);\r\n \/\/ ...\r\n int stepOver();\r\n int stepInto();\r\n int stepContinue();\r\n \/\/ ...\r\n SourceFile *_lastSource;\r\nprivate:\r\n bool compiledExists(Common::String filename);\r\n void reset();\r\n WintermuteEngine *_engine;\r\n int32 _lastDepth;\r\n ScScript *_lastScript;\r\n int32 _lastLine;\r\n};\r\n}<\/pre>\nAn example of how it works:<\/p>\n
\/\/ in debugger.cpp:\r\n \r\nbool Console::Cmd_AddBreakpoint(int argc, const char **argv) {\r\n \/**\r\n * Add a breakpoint\r\n *\/\r\n if (argc == 3) {\r\n int error = ADAPTER->addBreakpoint(argv[1], atoi(argv[2]));\r\n if (!error) {\r\n DebugPrintf(\"%s: OK\\n\", argv[0]);\r\n } else if (error == NO_SUCH_SCRIPT) {\r\n Common::String msg = Common::String::format(\"no such script: %s, breakpoint NOT created\\n\", argv[1]);\r\n debugWarning(argv[0], ERROR, msg);\r\n } else if (error == NO_SUCH_SOURCE) {\r\n Common::String msg = Common::String::format(\"no such source file: %s\\n\", argv[1]);\r\n debugWarning(argv[0], WARNING, msg);\r\n } else if (error == NO_SUCH_LINE) {\r\n Common::String msg = Common::String::format(\"source %s has no line %d\\n\", argv[1], atoi(argv[2]));\r\n debugWarning(argv[0], WARNING, msg);\r\n } else if (error == IS_BLANK) {\r\n Common::String msg = Common::String::format(\"%s:%d looks like a comment\/blank line.\\n\", argv[1], atoi(argv[2]));\r\n debugWarning(argv[0], WARNING, msg);\r\n } else {\r\n Common::String msg = Common::String::format(\"Error code %d\", error);\r\n debugWarning(argv[0], WARNING, msg);\r\n }\r\n } else {\r\n DebugPrintf(\"Usage: %s to break at line of file \\n\", argv[0]);\r\n }\r\n return true;\r\n}\r\n\r\n\/\/ in debugger_adapter.cpp\r\n \r\nint DebuggerAdapter::addBreakpoint(const char *filename, int line) {\r\n \/\/ TODO: Check if file exists, check if line exists\r\n assert(SCENGINE);\r\n if (!compiledExists(filename)) {\r\n return NO_SUCH_SCRIPT;\r\n }\r\n int isLegal = isBreakpointLegal(filename, line);\r\n if (isLegal == OK) {\r\n SCENGINE->addBreakpoint(filename, line);\r\n return OK;\r\n } else if (isLegal == IS_BLANK) {\r\n \/\/ We don't have the SOURCE. A warning will do.\r\n SCENGINE->addBreakpoint(filename, line);\r\n return IS_BLANK;\r\n } else if (isLegal == NO_SUCH_SOURCE) {\r\n \/\/ We don't have the SOURCE. A warning will do.\r\n SCENGINE->addBreakpoint(filename, line);\r\n return NO_SUCH_SOURCE;\r\n } else if (isLegal == NO_SUCH_LINE) {\r\n \/\/ No line in the source A warning will do.\r\n SCENGINE->addBreakpoint(filename, line);\r\n return NO_SUCH_LINE;\r\n } else {\r\n \/\/ Something weird? Don't do anything.\r\n return isLegal;\r\n }\r\n}\r\n<\/pre>\nThis is a somewhat trivial case since SCENGINE gives us addBreakpoint and we simply wrap it, but DebuggerAdapter does get more complex.<\/p>\n
This way, when and if we want a prettier user interface, we don\u2019t need to worry (not too much, at least) about breaking functionality, and if we decide to restructure the script engine and \/ or the way the adapter interacts with it, we don\u2019t need to worry too much about rendering the interface unusable.<\/p>\n
This could actually turn out useful if somebody decides to stitch together a graphical debugger \u2013 just code the actual GUI and add water.<\/p>\n
One cool thing our interface does is displaying source files \u2013 we\u2019ll talk about it in the next installment.<\/p>\n
<\/p>\n","protected":false},"excerpt":{"rendered":"
So, we were talking about the debugger. We\u2019ve seen how hooks are inserted into the execution unit, let\u2019s see how you interact with the debugger. We extend Common::Console, which already provides some very minimal functionality. It mainly enables us to easily bind commands to functions, tokenizing arguments on our behalf and giving us C-style argc\/argv. […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-57","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/posts\/57","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/comments?post=57"}],"version-history":[{"count":2,"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/posts\/57\/revisions"}],"predecessor-version":[{"id":61,"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/posts\/57\/revisions\/61"}],"wp:attachment":[{"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/media?parent=57"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/categories?post=57"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.scummvm.org\/t0by\/wp-json\/wp\/v2\/tags?post=57"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}