update wcclrun with variable dump mode, table header and some cleanups

wcclrun/main.cpp

 #include <cstdlib>
+#include <cstdio>
 #include <fstream>
 #include <iomanip>
 
@@ -11,7 +12,9 @@
 #include <boost/algorithm/string.hpp>
 #include <boost/make_shared.hpp>
 #include <boost/program_options.hpp>
+#include <boost/filesystem.hpp>
 #include <libcorpus2/io/xcesreader.h>
+#include <boost/lexical_cast.hpp>
 
 #include <antlr/NoViableAltException.hpp>
 #include <antlr/MismatchedTokenException.hpp>
@@ -20,6 +23,7 @@ namespace {
 	bool quiet = false;
 	bool tabs = false;
 	bool output_orths = true;
+	bool output_variables = false;
 	bool global_numbering = false;
 	bool in_sentence_numbering = true;
 }
@@ -54,6 +58,7 @@ private:
 	const Corpus2::Tagset& tagset_;
 	Wccl::Parser parser_;
 	std::vector< boost::shared_ptr<Wccl::FunctionalOperator> > ops_;
+	std::vector< std::string > op_names_;
 	int token_idx;
 };
 
@@ -68,6 +73,8 @@ bool Runner::load_more_operators(const std::string& filename)
 
 		retOp = parser_.parseAnyOperator(is);
 		if (retOp) {
+			boost::filesystem::path p(filename);
+			op_names_.push_back(p.stem());
 			ops_.push_back(retOp);
 			return true;
 		} else {
@@ -97,51 +104,76 @@ bool Runner::load_more_operators(const std::string& filename)
 void Runner::do_sentence(const boost::shared_ptr<Corpus2::Sentence>& sentence)
 {
 	Wccl::SentenceContext sc(sentence);
-	std::vector< std::vector< UnicodeString > > outputs(sentence->size());
-	std::vector<int> lengths(ops_.size() + 1, 0);
+	std::vector< std::vector< UnicodeString > > outputs;
+
 	streamsave sv(std::cout);
-	for (size_t i = 0; i < sentence->size(); ++i) {
-		sc.set_position(i);
-		UnicodeString orth = sentence->tokens()[i]->orth();
-		outputs[i].push_back(orth);
-		lengths[0] = std::max(lengths[0], orth.length());
-		int li = 1;
-		foreach (const boost::shared_ptr<Wccl::FunctionalOperator>& o, ops_) {
-			boost::shared_ptr<const Wccl::Value> v = o->base_apply(sc);
-			UnicodeString vstr = v->to_string_u(tagset_);
-			lengths[li] = std::max(lengths[li], vstr.length());
-			++li;
-			outputs[i].push_back(vstr);
+
+	if (1) { //header
+		outputs.resize(outputs.size() + 1);
+		std::vector< UnicodeString >& out = outputs.back();
+		if (global_numbering) {
+			out.push_back(UnicodeString::fromUTF8("##"));
+		}
+		if (in_sentence_numbering) {
+			out.push_back(UnicodeString::fromUTF8("#"));
+		}
+		if (output_orths) {
+			out.push_back(UnicodeString::fromUTF8("orth"));
+		}
+		for (size_t i = 0; i < op_names_.size(); ++i) {
+			out.push_back(UnicodeString::fromUTF8(op_names_[i]));
+			if (output_variables) {
+				boost::shared_ptr<Wccl::FunctionalOperator> o = ops_[i];
+				foreach (const std::string& varname, o->valid_variable_names()) {
+					const Wccl::Value& value = (*o)[varname];
+					std::string label = "(" + op_names_[i] + ")" + value.make_var_repr(varname);
+					out.push_back(UnicodeString::fromUTF8(label));
+				}
+			}
 		}
 	}
 	for (size_t i = 0; i < sentence->size(); ++i) {
+		outputs.resize(outputs.size() + 1);
+		std::vector< UnicodeString >& out = outputs.back();
+
 		++token_idx;
 		if (global_numbering) {
-			if (tabs) {
-				std::cout << token_idx << "\t";
-			} else {
-				std::cout << std::setw(6) << token_idx << " ";
-			}
+			out.push_back(UnicodeString::fromUTF8(boost::lexical_cast<std::string>(token_idx)));
 		}
 		if (in_sentence_numbering) {
-			if (tabs) {
-				std::cout << (i + 1) << "\t";
-			} else {
-				std::cout << std::setw(2) << (i + 1) << " ";
-			}
+			out.push_back(UnicodeString::fromUTF8(boost::lexical_cast<std::string>(i + 1)));
 		}
-		size_t b = output_orths ? 0 : 1;
-		for (size_t oi = b; oi < outputs[i].size(); ++oi) {
-			UnicodeString u = outputs[i][oi];
-			if (oi + 1 < outputs[i].size()) {
-				if (tabs) {
-					std::cout << PwrNlp::to_utf8(u);
-					std::cout << "\t";
-				} else {
-					u.padTrailing(lengths[oi] + 1);
-					std::cout << PwrNlp::to_utf8(u);
+		if (output_orths) {
+			out.push_back(sentence->tokens()[i]->orth());
+		}
+
+		sc.set_position(i);
+		foreach (const boost::shared_ptr<Wccl::FunctionalOperator>& o, ops_) {
+			boost::shared_ptr<const Wccl::Value> v = o->base_apply(sc);
+			UnicodeString vstr = v->to_string_u(tagset_);
+			out.push_back(vstr);
+			if (output_variables) {
+				foreach (const std::string& varname, o->valid_variable_names()) {
+					out.push_back((*o)[varname].to_string_u(tagset_));
 				}
+			}
+		}
+	}
+
+	std::vector<int> lengths(outputs[0].size());
+	foreach (const std::vector< UnicodeString >& line, outputs) {
+		for (size_t i = 0; i < line.size(); ++i) {
+			lengths[i] = std::max(lengths[i], line[i].length());
+		}
+	}
+
+	foreach (const std::vector< UnicodeString >& line, outputs) {
+		for (size_t i = 0; i < line.size(); ++i) {
+			UnicodeString u = line[i];
+			if (tabs) {
+				std::cout << PwrNlp::to_utf8(line[i]) << "\t";
 			} else {
+				u.padTrailing(lengths[i] + 1);
 				std::cout << PwrNlp::to_utf8(u);
 			}
 		}
@@ -192,6 +224,8 @@ int main(int argc, char** argv)
 			 "Output global counts\n")
 			("output-orths,O", value(&output_orths),
 			 "Output token orths\n")
+			("output-variables,V", value(&output_variables),
+			 "Output operator variables\n")
 			("help,h", "Show help")
 			;
 	boost::program_options::variables_map vm;