X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2FBiblioInfo.cpp;h=82690449ec33b5b558b086073c40c9333aff02c6;hb=ec3aed75ad965160a96547b7e36c924c5df7a91b;hp=99e3bd91acbdc80fccbf240b7fa2dd2b4251a858;hpb=206ca457517a7d6a2c92c23cf59f9e59241fb698;p=lyx.git diff --git a/src/BiblioInfo.cpp b/src/BiblioInfo.cpp index 99e3bd91ac..82690449ec 100644 --- a/src/BiblioInfo.cpp +++ b/src/BiblioInfo.cpp @@ -18,24 +18,20 @@ #include "buffer_funcs.h" #include "Encoding.h" #include "InsetIterator.h" +#include "Language.h" #include "Paragraph.h" +#include "TextClass.h" #include "TocBackend.h" -#include "insets/Inset.h" -#include "insets/InsetBibitem.h" -#include "insets/InsetBibtex.h" -#include "insets/InsetInclude.h" - #include "support/convert.h" #include "support/debug.h" #include "support/docstream.h" #include "support/gettext.h" #include "support/lassert.h" #include "support/lstrings.h" +#include "support/regex.h" #include "support/textutils.h" -#include "boost/regex.hpp" - #include using namespace std; @@ -86,7 +82,7 @@ docstring familyName(docstring const & name) for (; it != en; ++it) { if (!first) retval += " "; - else + else first = false; retval += *it; } @@ -113,7 +109,7 @@ docstring convertLaTeXCommands(docstring const & str) escaped = false; else if (ch == '\\') escaped = true; - else if (ch == '$') + else if (ch == '$') scanning_math = false; ret += ch; val = val.substr(1); @@ -173,14 +169,14 @@ docstring convertLaTeXCommands(docstring const & str) // unicodesymbols has things in the form: \"{u}, // whereas we may see things like: \"u. So we'll // look for that and change it, if necessary. - static boost::regex const reg("^\\\\\\W\\w"); - if (boost::regex_search(to_utf8(val), reg)) { + static lyx::regex const reg("^\\\\\\W\\w"); + if (lyx::regex_search(to_utf8(val), reg)) { val.insert(3, from_ascii("}")); val.insert(2, from_ascii("{")); } docstring rem; - docstring const cnvtd = Encodings::fromLaTeXCommand(val, rem, - Encodings::TEXT_CMD); + docstring const cnvtd = Encodings::fromLaTeXCommand(val, + Encodings::TEXT_CMD, rem); if (!cnvtd.empty()) { // it did, so we'll take that bit and proceed with what's left ret += cnvtd; @@ -220,7 +216,7 @@ docstring const BibTeXInfo::getAbbreviatedAuthor() const docstring authors; docstring const remainder = trim(split(opt, authors, '(')); if (remainder.empty()) - // in this case, we didn't find a "(", + // in this case, we didn't find a "(", // so we don't have author (year) return docstring(); return authors; @@ -235,7 +231,7 @@ docstring const BibTeXInfo::getAbbreviatedAuthor() const // FIXME Move this to a separate routine that can // be called from elsewhere. - // + // // OK, we've got some names. Let's format them. // Try to split the author list on " and " vector const authors = @@ -254,7 +250,7 @@ docstring const BibTeXInfo::getAbbreviatedAuthor() const docstring const BibTeXInfo::getYear() const { - if (is_bibtex_) + if (is_bibtex_) return operator[]("year"); docstring const opt = label(); @@ -263,7 +259,7 @@ docstring const BibTeXInfo::getYear() const docstring authors; docstring tmp = split(opt, authors, '('); - if (tmp.empty()) + if (tmp.empty()) // we don't have author (year) return docstring(); docstring year; @@ -281,22 +277,21 @@ docstring const BibTeXInfo::getXRef() const namespace { - docstring parseOptions(docstring const & format, docstring & optkey, - docstring & ifpart, docstring & elsepart); - - /// Calls parseOptions to deal with an embedded option, such as: - /// {%number%[[, no.~%number%]]} - /// which must appear at the start of format. ifelsepart gets the - /// whole of the option, and we return what's left after the option. - /// we return format if there is an error. - docstring parseEmbeddedOption(docstring const & format, - docstring & ifelsepart) + string parseOptions(string const & format, string & optkey, + string & ifpart, string & elsepart); + + // Calls parseOptions to deal with an embedded option, such as: + // {%number%[[, no.~%number%]]} + // which must appear at the start of format. ifelsepart gets the + // whole of the option, and we return what's left after the option. + // we return format if there is an error. + string parseEmbeddedOption(string const & format, string & ifelsepart) { LASSERT(format[0] == '{' && format[1] == '%', return format); - docstring optkey; - docstring ifpart; - docstring elsepart; - docstring const rest = parseOptions(format, optkey, ifpart, elsepart); + string optkey; + string ifpart; + string elsepart; + string const rest = parseOptions(format, optkey, ifpart, elsepart); if (format == rest) { // parse error LYXERR0("ERROR! Couldn't parse `" << format <<"'."); return format; @@ -305,30 +300,30 @@ namespace { ifelsepart = format.substr(0, format.size() - rest.size()); return rest; } - - - // Gets a "clause" from a format string, where the clause is + + + // Gets a "clause" from a format string, where the clause is // delimited by '[[' and ']]'. Returns what is left after the // clause is removed, and returns format if there is an error. - docstring getClause(docstring const & format, docstring & clause) + string getClause(string const & format, string & clause) { - docstring fmt = format; + string fmt = format; // remove '[[' fmt = fmt.substr(2); - // we'll remove characters from the front of fmt as we + // we'll remove characters from the front of fmt as we // deal with them - while (fmt.size()) { + while (fmt.size()) { if (fmt[0] == ']' && fmt.size() > 1 && fmt[1] == ']') { - // that's the end + // that's the end fmt = fmt.substr(2); break; } // check for an embedded option if (fmt[0] == '{' && fmt.size() > 1 && fmt[1] == '%') { - docstring part; - docstring const rest = parseEmbeddedOption(fmt, part); + string part; + string const rest = parseEmbeddedOption(fmt, part); if (fmt == rest) { - LYXERR0("ERROR! Couldn't parse `" << format <<"'."); + LYXERR0("ERROR! Couldn't parse embedded option in `" << format <<"'."); return format; } clause += part; @@ -342,16 +337,16 @@ namespace { } - /// parse an options string, which must appear at the start of the - /// format parameter. puts the parsed bits in optkey, ifpart, and - /// elsepart and returns what's left after the option is removed. - /// if there's an error, it returns format itself. - docstring parseOptions(docstring const & format, docstring & optkey, - docstring & ifpart, docstring & elsepart) + // parse an options string, which must appear at the start of the + // format parameter. puts the parsed bits in optkey, ifpart, and + // elsepart and returns what's left after the option is removed. + // if there's an error, it returns format itself. + string parseOptions(string const & format, string & optkey, + string & ifpart, string & elsepart) { LASSERT(format[0] == '{' && format[1] == '%', return format); // strip '{%' - docstring fmt = format.substr(2); + string fmt = format.substr(2); size_t pos = fmt.find('%'); // end of key if (pos == string::npos) { LYXERR0("Error parsing `" << format <<"'. Can't find end of key."); @@ -365,7 +360,7 @@ namespace { return format; } - docstring curfmt = fmt; + string curfmt = fmt; fmt = getClause(curfmt, ifpart); if (fmt == curfmt) { LYXERR0("Error parsing `" << format <<"'. Couldn't get if clause."); @@ -374,13 +369,13 @@ namespace { if (fmt[0] == '}') // we're done, no else clause return fmt.substr(1); - + // else part should follow if (fmt[0] != '[' || fmt[1] != '[') { LYXERR0("Error parsing `" << format <<"'. Can't find else clause."); return format; } - + curfmt = fmt; fmt = getClause(curfmt, elsepart); // we should be done @@ -394,91 +389,134 @@ namespace { } // anon namespace -docstring BibTeXInfo::expandFormat(docstring const & format, - BibTeXInfo const * const xref) const +docstring BibTeXInfo::expandFormat(string const & format, + BibTeXInfo const * const xref, int & counter, Buffer const & buf, + bool richtext) const { - // return value - docstring ret; - docstring key; + // incorrect use of macros could put us in an infinite loop + static int max_passes = 5000; + docstring ret; // return value + string key; bool scanning_key = false; + bool scanning_rich = false; - docstring fmt = format; - // we'll remove characters from the front of fmt as we + string fmt = format; + // we'll remove characters from the front of fmt as we // deal with them while (fmt.size()) { + if (counter++ > max_passes) { + LYXERR0("Recursion limit reached while parsing `" + << format << "'."); + return _("ERROR!"); + } + char_type thischar = fmt[0]; - if (thischar == '%') { + if (thischar == '%') { // beginning or end of key - if (scanning_key) { + if (scanning_key) { // end of key scanning_key = false; // so we replace the key with its value, which may be empty - docstring const val = getValueForKey(to_utf8(key), xref); - key.clear(); - ret += val; + if (key[0] == '!') { + // macro + string const val = + buf.params().documentClass().getCiteMacro(key); + fmt = val + fmt.substr(1); + continue; + } else if (key[0] == '_') { + // a translatable bit + string const val = + buf.params().documentClass().getCiteMacro(key); + docstring const trans = + translateIfPossible(from_utf8(val), buf.params().language->code()); + ret += trans; + } else { + docstring const val = getValueForKey(key, xref); + ret += val; + } } else { // beginning of key + key.clear(); scanning_key = true; } - } - else if (thischar == '{') { + } + else if (thischar == '{') { // beginning of option? if (scanning_key) { LYXERR0("ERROR: Found `{' when scanning key in `" << format << "'."); return _("ERROR!"); } - if (fmt.size() > 1 && fmt[1] == '%') { - // it is the beginning of an optional format - docstring optkey; - docstring ifpart; - docstring elsepart; - docstring const newfmt = + if (fmt.size() > 1) { + if (fmt[1] == '%') { + // it is the beginning of an optional format + string optkey; + string ifpart; + string elsepart; + string const newfmt = parseOptions(fmt, optkey, ifpart, elsepart); - if (newfmt == fmt) // parse error - return _("ERROR!"); - fmt = newfmt; - docstring const val = getValueForKey(to_utf8(optkey), xref); - if (!val.empty()) - ret += expandFormat(ifpart, xref); - else if (!elsepart.empty()) - ret += expandFormat(elsepart, xref); - // fmt will have been shortened for us already - continue; - } + if (newfmt == fmt) // parse error + return _("ERROR!"); + fmt = newfmt; + docstring const val = getValueForKey(optkey, xref); + if (!val.empty()) + ret += expandFormat(ifpart, xref, counter, buf, richtext); + else if (!elsepart.empty()) + ret += expandFormat(elsepart, xref, counter, buf, richtext); + // fmt will have been shortened for us already + continue; + } + if (fmt[1] == '!') { + // beginning of rich text + scanning_rich = true; + fmt = fmt.substr(2); + continue; + } + } + // we are here if '{' was not followed by % or !. + // So it's just a character. ret += thischar; - } + } + else if (scanning_rich && thischar == '!' + && fmt.size() > 1 && fmt[1] == '}') { + // end of rich text + scanning_rich = false; + fmt = fmt.substr(2); + continue; + } else if (scanning_key) - key += thischar; - else + key += char(thischar); + else if (richtext) { + if (scanning_rich) + ret += thischar; + else { + // we need to escape '<' and '>' + if (thischar == '<') + ret += "<"; + else if (thischar == '>') + ret += ">"; + else + ret += thischar; + } + } else if (!scanning_rich /* && !richtext */) ret += thischar; + // else the character is discarded, which will happen only if + // richtext == false and we are scanning rich text fmt = fmt.substr(1); } // for loop if (scanning_key) { LYXERR0("Never found end of key in `" << format << "'!"); return _("ERROR!"); } + if (scanning_rich) { + LYXERR0("Never found end of rich text in `" << format << "'!"); + return _("ERROR!"); + } return ret; } -namespace { - -// FIXME These would be better read from a file, so that they -// could be customized. - - static docstring articleFormat = from_ascii("%author%, \"%title%\", %journal% {%volume%[[ %volume%{%number%[[, %number%]]}]]} (%year%){%pages%[[, pp. %pages%]]}.{%note%[[ %note%]]}"); - -static docstring bookFormat = from_ascii("{%author%[[%author%]][[%editor%, ed.]]}, %title%{%volume%[[ vol. %volume%]][[{%number%[[no. %number%]]}]]}{%edition%[[%edition%]]} ({%address%[[%address%: ]]}%publisher%, %year%).{%note%[[ %note%]]}"); - -static docstring inSomething = from_ascii("%author%, \"%title%\", in{%editor%[[ %editor%, ed.,]]} %booktitle%{%volume%[[ vol. %volume%]][[{%number%[[no. %number%]]}]]}{%edition%[[%edition%]]} ({%address%[[%address%: ]]}%publisher%, %year%){%pages%[[, pp. %pages%]]}.{%note%[[ %note%]]}"); - -static docstring thesis = from_ascii("%author%, %title% ({%address%[[%address%: ]]}%school%, %year%).{%note%[[ %note%]]}"); - -static docstring defaultFormat = from_ascii("{%author%[[%author%, ]][[{%editor%[[%editor%, ed., ]]}]]}%title%{%journal%[[, %journal%]][[{%publisher%[[, %publisher%]][[{%institution%[[, %institution%]]}]]}]]}{%year%[[ (%year%)]]}{%pages%[[, %pages%]]}."); - -} - -docstring const & BibTeXInfo::getInfo(BibTeXInfo const * const xref) const +docstring const & BibTeXInfo::getInfo(BibTeXInfo const * const xref, + Buffer const & buf, bool richtext) const { if (!info_.empty()) return info_; @@ -489,16 +527,10 @@ docstring const & BibTeXInfo::getInfo(BibTeXInfo const * const xref) const return info_; } - if (entry_type_ == "article") - info_ = expandFormat(articleFormat, xref); - else if (entry_type_ == "book") - info_ = expandFormat(bookFormat, xref); - else if (entry_type_.substr(0,2) == "in") - info_ = expandFormat(inSomething, xref); - else if (entry_type_ == "phdthesis" || entry_type_ == "mastersthesis") - info_ = expandFormat(thesis, xref); - else - info_ = expandFormat(defaultFormat, xref); + DocumentClass const & dc = buf.params().documentClass(); + string const & format = dc.getCiteFormat(to_utf8(entry_type_)); + int counter = 0; + info_ = expandFormat(format, xref, counter, buf, richtext); if (!info_.empty()) info_ = convertLaTeXCommands(info_); @@ -514,15 +546,15 @@ docstring const & BibTeXInfo::operator[](docstring const & field) const static docstring const empty_value = docstring(); return empty_value; } - - + + docstring const & BibTeXInfo::operator[](string const & field) const { return operator[](from_ascii(field)); } -docstring BibTeXInfo::getValueForKey(string const & key, +docstring BibTeXInfo::getValueForKey(string const & key, BibTeXInfo const * const xref) const { docstring const ret = operator[](key); @@ -629,11 +661,12 @@ docstring const BiblioInfo::getYear(docstring const & key, bool use_modifier) co } -docstring const BiblioInfo::getInfo(docstring const & key) const +docstring const BiblioInfo::getInfo(docstring const & key, + Buffer const & buf, bool richtext) const { BiblioInfo::const_iterator it = find(key); if (it == end()) - return docstring(); + return docstring(_("Bibliography entry not found!")); BibTeXInfo const & data = it->second; BibTeXInfo const * xrefptr = 0; docstring const xref = data.getXRef(); @@ -642,7 +675,7 @@ docstring const BiblioInfo::getInfo(docstring const & key) const if (xrefit != end()) xrefptr = &(xrefit->second); } - return data.getInfo(xrefptr); + return data.getInfo(xrefptr, buf, richtext); } @@ -659,8 +692,8 @@ bool BiblioInfo::isBibtex(docstring const & key) const vector const BiblioInfo::getCiteStrings( docstring const & key, Buffer const & buf) const { - CiteEngine const engine = buf.params().citeEngine(); - if (engine == ENGINE_BASIC || engine == ENGINE_NATBIB_NUMERICAL) + CiteEngineType const engine_type = buf.params().citeEngineType(); + if (engine_type == ENGINE_TYPE_NUMERICAL) return getNumericalStrings(key, buf); else return getAuthorYearStrings(key, buf); @@ -678,8 +711,9 @@ vector const BiblioInfo::getNumericalStrings( if (author.empty() || year.empty()) return vector(); - vector const & styles = citeStyles(buf.params().citeEngine()); - + vector const & styles = citeStyles(buf.params().citeEngine(), + buf.params().citeEngineType()); + vector vec(styles.size()); for (size_t i = 0; i != vec.size(); ++i) { docstring str; @@ -737,8 +771,9 @@ vector const BiblioInfo::getAuthorYearStrings( if (author.empty() || year.empty()) return vector(); - vector const & styles = citeStyles(buf.params().citeEngine()); - + vector const & styles = citeStyles(buf.params().citeEngine(), + buf.params().citeEngineType()); + vector vec(styles.size()); for (size_t i = 0; i != vec.size(); ++i) { docstring str; @@ -791,6 +826,8 @@ vector const BiblioInfo::getAuthorYearStrings( void BiblioInfo::mergeBiblioInfo(BiblioInfo const & info) { bimap_.insert(info.begin(), info.end()); + field_names_.insert(info.field_names_.begin(), info.field_names_.end()); + entry_types_.insert(info.entry_types_.begin(), info.entry_types_.end()); } @@ -830,9 +867,9 @@ void BiblioInfo::collectCitedEntries(Buffer const & buf) } if (citekeys.empty()) return; - + // We have a set of the keys used in this document. - // We will now convert it to a list of the BibTeXInfo objects used in + // We will now convert it to a list of the BibTeXInfo objects used in // this document... vector bi; set::const_iterator cit = citekeys.begin(); @@ -845,7 +882,7 @@ void BiblioInfo::collectCitedEntries(Buffer const & buf) } // ...and sort it. sort(bi.begin(), bi.end(), lSorter); - + // Now we can write the sorted keys vector::const_iterator bit = bi.begin(); vector::const_iterator ben = bi.end(); @@ -857,9 +894,8 @@ void BiblioInfo::collectCitedEntries(Buffer const & buf) void BiblioInfo::makeCitationLabels(Buffer const & buf) { collectCitedEntries(buf); - CiteEngine const engine = buf.params().citeEngine(); - bool const numbers = - (engine == ENGINE_BASIC || engine == ENGINE_NATBIB_NUMERICAL); + CiteEngineType const engine_type = buf.params().citeEngineType(); + bool const numbers = (engine_type == ENGINE_TYPE_NUMERICAL); int keynumber = 0; char modifier = 0; @@ -897,7 +933,7 @@ void BiblioInfo::makeCitationLabels(Buffer const & buf) } else { modifier = 0; } - entry.setModifier(modifier); + entry.setModifier(modifier); // remember the last one last = biit; } @@ -990,32 +1026,47 @@ string citationStyleToString(const CitationStyle & s) return cite; } -vector citeStyles(CiteEngine engine) +vector citeStyles(CiteEngine engine, CiteEngineType engine_type) { - unsigned int nStyles = 0; - unsigned int start = 0; + vector styles(0); - switch (engine) { + if (engine_type == ENGINE_TYPE_AUTHORYEAR) { + switch (engine) { case ENGINE_BASIC: - nStyles = 2; - start = 0; + styles.push_back(CITE); + break; + case ENGINE_JURABIB: + styles.push_back(CITE); + case ENGINE_NATBIB: + styles.push_back(CITET); + styles.push_back(CITEP); + styles.push_back(CITEALT); + styles.push_back(CITEALP); + styles.push_back(CITEAUTHOR); + styles.push_back(CITEYEAR); + styles.push_back(CITEYEARPAR); break; - case ENGINE_NATBIB_AUTHORYEAR: - case ENGINE_NATBIB_NUMERICAL: - nStyles = nCiteStyles - 1; - start = 1; + } + } else { + switch (engine) { + case ENGINE_BASIC: + styles.push_back(CITE); break; case ENGINE_JURABIB: - nStyles = nCiteStyles; - start = 0; + styles.push_back(CITE); + case ENGINE_NATBIB: + styles.push_back(CITET); + styles.push_back(CITEALT); + styles.push_back(CITEAUTHOR); + styles.push_back(CITEP); + styles.push_back(CITEALP); + styles.push_back(CITEYEAR); + styles.push_back(CITEYEARPAR); break; + } } - vector styles(nStyles); - size_t i = 0; - int j = start; - for (; i != styles.size(); ++i, ++j) - styles[i] = citeStylesArray[j]; + styles.push_back(NOCITE); return styles; }