#include <config.h>
-#include "Encoding.h"
#include "Parser.h"
+
+#include "Encoding.h"
#include "support/lstrings.h"
#include "support/textutils.h"
}
-void iparserdocstream::putback(docstring s)
+void iparserdocstream::putback(docstring const & s)
{
s_ = s + s_;
}
Parser::Parser(idocstream & is, std::string const & fixedenc)
- : lineno_(0), pos_(0), iss_(0), is_(is),
+ : lineno_(0), pos_(0), iss_(nullptr), is_(is),
encoding_iconv_(fixedenc.empty() ? "UTF-8" : fixedenc),
theCatcodesType_(NORMAL_CATCODES), curr_cat_(UNDECIDED_CATCODES),
fixed_enc_(!fixedenc.empty())
}
-bool Parser::setEncoding(std::string const & e, int const & p)
+bool Parser::setEncoding(std::string const & e, int p)
{
// We may (and need to) use unsafe encodings here: Since the text is
// converted to unicode while reading from is_, we never see text in
bool Parser::isParagraph()
{
- // A new paragraph in TeX ist started
+ // A new paragraph in TeX is started
// - either by a newline, following any amount of whitespace
// characters (including zero), and another newline
// - or the token \par
}
-bool Parser::good()
+bool Parser::good() const
{
if (pos_ < tokens_.size())
return true;
}
-bool Parser::hasOpt()
+bool Parser::hasOpt(string const & l)
{
// An optional argument can occur in any of the following forms:
// - \foo[bar]
putback();
break;
}
- bool const retval = (next_token().asInput() == "[");
+ bool const retval = (next_token().asInput() == l);
pos_ = oldpos;
return retval;
}
if (! good())
return make_pair(false, string());
+ int group_level = 0;
string result;
Token t = get_token();
} else {
while (good()) {
t = get_token();
+ // honor grouping
+ if (left != '{' && t.cat() == catBegin) {
+ ++group_level;
+ continue;
+ }
+ if (left != '{' && t.cat() == catEnd) {
+ --group_level;
+ continue;
+ }
// Ignore comments
if (t.cat() == catComment) {
if (!t.cs().empty())
continue;
}
if (allow_escaping) {
- if (t.cat() != catEscape && t.character() == right)
+ if (t.cat() != catEscape && t.character() == right
+ && group_level == 0)
break;
} else {
if (t.character() == right) {
if (t.cat() == catEscape)
result += '\\';
- break;
+ if (group_level == 0)
+ break;
}
}
result += t.asInput();
}
+bool Parser::hasListPreamble(string const & itemcmd)
+{
+ // remember current position
+ unsigned int oldpos = pos_;
+ // jump over arguments
+ if (hasOpt())
+ getOpt();
+ if (hasOpt("{"))
+ getArg('{', '}');
+ // and swallow spaces and comments
+ skip_spaces(true);
+ // we have a list preamble if the next thing
+ // that follows is not the \item command
+ bool res = next_token().cs() != itemcmd;
+ // back to orig position
+ pos_ = oldpos;
+ return res;
+}
+
+
string const Parser::ertEnvironment(string const & name)
{
if (!good())
}
+string const Parser::getCommandLatexParam()
+{
+ if (!good())
+ return string();
+ string res;
+ size_t offset = 0;
+ while (true) {
+ if (pos_ + offset >= tokens_.size())
+ tokenize_one();
+ if (pos_ + offset >= tokens_.size())
+ break;
+ Token t = tokens_[pos_ + offset];
+ if (t.cat() == catBegin)
+ break;
+ res += t.asInput();
+ ++offset;
+ }
+ return res;
+}
+
+
Parser::Arg Parser::verbatimStuff(string const & end_string, bool const allow_linebreak)
{
if (!good())
}
case catIgnore: {
- cerr << "ignoring a char: " << c << "\n";
+ cerr << "ignoring a char: " << static_cast<uint32_t>(c) << "\n";
break;
}
}
-void Parser::error(string const & msg)
+void Parser::error(string const & msg) const
{
cerr << "Line ~" << lineno_ << ": parse error: " << msg << endl;
dump();