-/* This file is part of
- * ======================================================
- *
- * LyX, The Document Processor
- *
- * Copyright 2001 The LyX Team.
+/**
+ * \file biblio.C
+ * This file is part of LyX, the document processor.
+ * Licence details can be found in the file COPYING.
*
- * ======================================================
+ * \author Angus Leeming
+ * \author Herbert Voss
*
- * \file biblio.C
- * \author Angus Leeming <a.leeming@ic.ac.uk>
- * \author Herbert Voss <voss@perce.de>
+ * Full author contact details are available in file CREDITS.
*/
#include <config.h>
-#include <vector>
-#include <algorithm>
-
-#ifdef __GNUG__
-#pragma implementation
-#endif
-
-#include "LString.h"
#include "biblio.h"
+
+#include "support/std_sstream.h"
#include "gettext.h" // for _()
-#include "helper_funcs.h"
+
#include "support/lstrings.h"
#include "support/LAssert.h"
-#include "support/LRegex.h"
-
-using std::find;
-using std::min;
-using std::vector;
-using std::sort;
-
-namespace biblio
-{
-
-namespace {
-
-using namespace biblio;
-
-char const * const citeCommands[] = {
- "cite", "citet", "citep", "citealt", "citealp", "citeauthor",
- "citeyear", "citeyearpar" };
-
-unsigned int const nCiteCommands =
- sizeof(citeCommands) / sizeof(char *);
-
-CiteStyle const citeStyles[] = {
- CITE, CITET, CITEP, CITEALT, CITEALP,
- CITEAUTHOR, CITEYEAR, CITEYEARPAR };
-
-unsigned int const nCiteStyles =
- sizeof(citeStyles) / sizeof(CiteStyle);
-
-CiteStyle const citeStylesFull[] = {
- CITET, CITEP, CITEALT, CITEALP, CITEAUTHOR };
-
-unsigned int const nCiteStylesFull =
- sizeof(citeStylesFull) / sizeof(CiteStyle);
-
-CiteStyle const citeStylesUCase[] = {
- CITET, CITEP, CITEALT, CITEALP, CITEAUTHOR };
-
-unsigned int const nCiteStylesUCase =
- sizeof(citeStylesUCase) / sizeof(CiteStyle);
-
-
-// The functions doing the dirty work for the search.
-vector<string>::const_iterator
-simpleSearch(InfoMap const & theMap,
- vector<string> const & keys,
- string const & expr,
- vector<string>::const_iterator start,
- Direction dir,
- bool caseSensitive)
-{
- string tmp = expr;
- if (!caseSensitive)
- tmp = lowercase(tmp);
-
- vector<string> searchwords = getVectorFromString(tmp, " ");
-
- // Loop over all keys from start...
- for (vector<string>::const_iterator it = start;
- // End condition is direction-dependent.
- (dir == FORWARD) ? (it<keys.end()) : (it>=keys.begin());
- // increment is direction-dependent.
- (dir == FORWARD) ? (++it) : (--it)) {
-
- string data = (*it);
- InfoMap::const_iterator info = theMap.find(*it);
- if (info != theMap.end())
- data += " " + info->second;
- if (!caseSensitive)
- data = lowercase(data);
-
- bool found = true;
-
- // Loop over all search words...
- for (vector<string>::const_iterator sit = searchwords.begin();
- sit != searchwords.end(); ++sit) {
- if (data.find(*sit) == string::npos) {
- found = false;
- break;
- }
- }
-
- if (found) return it;
- }
-
- return keys.end();
-}
-
-
-vector<string>::const_iterator
-regexSearch(InfoMap const & theMap,
- vector<string> const & keys,
- string const & expr,
- vector<string>::const_iterator start,
- Direction dir)
-{
- LRegex reg(expr);
-
- for (vector<string>::const_iterator it = start;
- // End condition is direction-dependent.
- (dir == FORWARD) ? (it<keys.end()) : (it>=keys.begin());
- // increment is direction-dependent.
- (dir == FORWARD) ? (++it) : (--it)) {
- string data = (*it);
- InfoMap::const_iterator info = theMap.find(*it);
- if (info != theMap.end())
- data += " " + info->second;
-
- if (reg.exec(data).size() > 0)
- return it;
- }
+#include <boost/regex.hpp>
- return keys.end();
-}
+using namespace lyx::support;
+using std::vector;
-} // namespace anon
+namespace biblio {
string const familyName(string const & name)
{
// "Surname, F."
// "FirstName Surname"
// "F. Surname"
- string::size_type idx = fname.find(",");
+ string::size_type idx = fname.find(',');
if (idx != string::npos)
- return frontStrip(fname.substr(0,idx));
- idx = fname.rfind(".");
+ return ltrim(fname.substr(0, idx));
+ idx = fname.rfind('.');
if (idx != string::npos)
- fname = frontStrip(fname.substr(idx+1));
+ fname = ltrim(fname.substr(idx + 1));
// test if we have a LaTeX Space in front
if (fname[0] == '\\')
return fname.substr(2);
- return fname;
+ return rtrim(fname);
}
string const getAbbreviatedAuthor(InfoMap const & map, string const & key)
{
- lyx::Assert(!map.empty());
+ Assert(!map.empty());
InfoMap::const_iterator it = map.find(key);
+ if (it == map.end())
+ return string();
+ string const & data = it->second;
- string author;
- if (it != map.end()) {
- author = parseBibTeX(it->second, "author");
- if (author.empty())
- author = parseBibTeX(it->second, "editor");
+ // Is the entry a BibTeX one or one from lyx-layout "bibliography"?
+ string::size_type const pos = data.find("TheBibliographyRef");
+ if (pos != string::npos) {
+ if (pos <= 2) {
+ return string();
+ }
- vector<string> authors = getVectorFromString(author, "and");
+ string const opt = trim(data.substr(0, pos - 1));
+ if (opt.empty())
+ return string();
- if (!authors.empty()) {
- author.erase();
+ string authors;
+ split(opt, authors, '(');
+ return authors;
+ }
- for (vector<string>::iterator it = authors.begin();
- it != authors.end(); ++it) {
- *it = familyName(strip(*it));
- }
+ string author = parseBibTeX(data, "author");
- author = authors[0];
- if (authors.size() == 2)
- author += _(" and ") + authors[1];
- else if (authors.size() > 2)
- author += _(" et al.");
- }
+ if (author.empty())
+ author = parseBibTeX(data, "editor");
+
+ if (author.empty()) {
+ author = parseBibTeX(data, "key");
+ if (author.empty())
+ author = key;
+ return author;
}
- if (author.empty())
- author = _("Caesar et al.");
+ vector<string> const authors = getVectorFromString(author, " and ");
+ if (authors.empty())
+ return author;
+
+ if (authors.size() == 2)
+ return bformat(_("%1$s and %2$s"),
+ familyName(authors[0]), familyName(authors[1]));
- return author;
+ if (authors.size() > 2)
+ return bformat(_("%1$s et al."), familyName(authors[0]));
+
+ return familyName(authors[0]);
}
string const getYear(InfoMap const & map, string const & key)
{
- lyx::Assert(!map.empty());
+ Assert(!map.empty());
InfoMap::const_iterator it = map.find(key);
+ if (it == map.end())
+ return string();
+ string const & data = it->second;
+
+ // Is the entry a BibTeX one or one from lyx-layout "bibliography"?
+ string::size_type const pos = data.find("TheBibliographyRef");
+ if (pos != string::npos) {
+ if (pos <= 2) {
+ return string();
+ }
- string year;
+ string const opt =
+ trim(data.substr(0, pos - 1));
+ if (opt.empty())
+ return string();
- if (it != map.end())
- year = parseBibTeX(it->second, "year");
+ string authors;
+ string const tmp = split(opt, authors, '(');
+ string year;
+ split(tmp, year, ')');
+ return year;
+ }
+
+ string year = parseBibTeX(data, "year");
if (year.empty())
- year = "50BC";
+ year = _("No year");
return year;
}
+namespace {
+
// A functor for use with std::sort, leading to case insensitive sorting
struct compareNoCase: public std::binary_function<string, string, bool>
{
bool operator()(string const & s1, string const & s2) const {
- return compare_no_case(s1, s2) < 0;
+ return compare_ascii_no_case(s1, s2) < 0;
}
};
+} // namespace anon
+
+
vector<string> const getKeys(InfoMap const & map)
{
vector<string> bibkeys;
-
- for (InfoMap::const_iterator it = map.begin(); it != map.end(); ++it) {
+ InfoMap::const_iterator it = map.begin();
+ InfoMap::const_iterator end = map.end();
+ for (; it != end; ++it) {
bibkeys.push_back(it->first);
}
- sort(bibkeys.begin(), bibkeys.end(), compareNoCase());
+ std::sort(bibkeys.begin(), bibkeys.end(), compareNoCase());
return bibkeys;
}
string const getInfo(InfoMap const & map, string const & key)
{
- lyx::Assert(!map.empty());
+ Assert(!map.empty());
InfoMap::const_iterator it = map.find(key);
if (it == map.end())
return string();
+ string const & data = it->second;
+
// is the entry a BibTeX one or one from lyx-layout "bibliography"?
- if (!contains(it->second,'='))
- return it->second.c_str();
+ string const separator("TheBibliographyRef");
+ string::size_type const pos = data.find(separator);
+ if (pos != string::npos) {
+ string::size_type const pos2 = pos + separator.size();
+ string const info = trim(data.substr(pos2));
+ return info;
+ }
// Search for all possible "required" keys
- string author = parseBibTeX(it->second, "author");
+ string author = parseBibTeX(data, "author");
if (author.empty())
- author = parseBibTeX(it->second, "editor");
+ author = parseBibTeX(data, "editor");
- string year = parseBibTeX(it->second, "year");
- string title = parseBibTeX(it->second, "title");
- string booktitle = parseBibTeX(it->second, "booktitle");
- string chapter = parseBibTeX(it->second, "chapter");
- string number = parseBibTeX(it->second, "number");
- string volume = parseBibTeX(it->second, "volume");
- string pages = parseBibTeX(it->second, "pages");
+ string year = parseBibTeX(data, "year");
+ string title = parseBibTeX(data, "title");
+ string booktitle = parseBibTeX(data, "booktitle");
+ string chapter = parseBibTeX(data, "chapter");
+ string number = parseBibTeX(data, "number");
+ string volume = parseBibTeX(data, "volume");
+ string pages = parseBibTeX(data, "pages");
- string media = parseBibTeX(it->second, "journal");
+ string media = parseBibTeX(data, "journal");
if (media.empty())
- media = parseBibTeX(it->second, "publisher");
+ media = parseBibTeX(data, "publisher");
if (media.empty())
- media = parseBibTeX(it->second, "school");
+ media = parseBibTeX(data, "school");
if (media.empty())
- media = parseBibTeX(it->second, "institution");
+ media = parseBibTeX(data, "institution");
ostringstream result;
if (!author.empty())
if (!year.empty())
result << ", " << year;
- string const result_str = strip(result.str().c_str());
+ string const result_str = rtrim(STRCONV(result.str()));
if (!result_str.empty())
return result_str;
// This should never happen (or at least be very unusual!)
- return it->second;
+ return data;
}
+namespace {
+
+// Escape special chars.
+// All characters are literals except: '.|*?+(){}[]^$\'
+// These characters are literals when preceded by a "\", which is done here
+string const escape_special_chars(string const & expr)
+{
+ // Search for all chars '.|*?+(){}[^$]\'
+ // Note that '[' and '\' must be escaped.
+ // This is a limitation of boost::regex, but all other chars in BREs
+ // are assumed literal.
+ boost::RegEx reg("[].|*?+(){}^$\\[\\\\]");
+
+ // $& is a perl-like expression that expands to all of the current match
+ // The '$' must be prefixed with the escape character '\' for
+ // boost to treat it as a literal.
+ // Thus, to prefix a matched expression with '\', we use:
+ return STRCONV(reg.Merge(STRCONV(expr), "\\\\$&"));
+}
+
+
+// A functor for use with std::find_if, used to ascertain whether a
+// data entry matches the required regex_
+struct RegexMatch
+{
+ // re and icase are used to construct an instance of boost::RegEx.
+ // if icase is true, then matching is insensitive to case
+ RegexMatch(InfoMap const & m, string const & re, bool icase)
+ : map_(m), regex_(STRCONV(re), icase) {}
+
+ bool operator()(string const & key) {
+ if (!validRE())
+ return false;
+
+ // the data searched is the key + its associated BibTeX/biblio
+ // fields
+ string data = key;
+ InfoMap::const_iterator info = map_.find(key);
+ if (info != map_.end())
+ data += ' ' + info->second;
+
+ // Attempts to find a match for the current RE
+ // somewhere in data.
+ return regex_.Search(STRCONV(data));
+ }
+
+ bool validRE() const { return regex_.error_code() == 0; }
+
+private:
+ InfoMap const map_;
+ boost::RegEx regex_;
+};
+
+} // namespace anon
+
+
vector<string>::const_iterator
searchKeys(InfoMap const & theMap,
vector<string> const & keys,
- string const & expr,
+ string const & search_expr,
vector<string>::const_iterator start,
Search type,
Direction dir,
if (start < keys.begin() || start >= keys.end())
return keys.end();
- string search_expr = frontStrip(strip(expr));
- if (search_expr.empty())
+ string expr = trim(search_expr);
+ if (expr.empty())
return keys.end();
if (type == SIMPLE)
- return simpleSearch(theMap, keys, search_expr, start, dir,
- caseSensitive);
+ // We must escape special chars in the search_expr so that
+ // it is treated as a simple string by boost::regex.
+ expr = escape_special_chars(expr);
- return regexSearch(theMap, keys, search_expr, start, dir);
+ // Build the functor that will be passed to find_if.
+ RegexMatch const match(theMap, expr, !caseSensitive);
+ if (!match.validRE())
+ return keys.end();
+
+ // Search the vector of 'keys' from 'start' for one that matches the
+ // predicate 'match'. Searching can be forward or backward from start.
+ if (dir == FORWARD)
+ return std::find_if(start, keys.end(), match);
+
+ vector<string>::const_reverse_iterator rit(start);
+ vector<string>::const_reverse_iterator rend = keys.rend();
+ rit = std::find_if(rit, rend, match);
+
+ if (rit == rend)
+ return keys.end();
+ // This is correct and always safe.
+ // (See Meyer's Effective STL, Item 28.)
+ return (++rit).base();
}
string keyvalue;
// at first we delete all characters right of '%' and
// replace tabs through a space and remove leading spaces
+ // we read the data line by line so that the \n are
+ // ignored, too.
string data_;
int Entries = 0;
string dummy = token(data,'\n', Entries);
while (!dummy.empty()) {
dummy = subst(dummy, '\t', ' '); // no tabs
- dummy = frontStrip(dummy); // no leading spaces
+ dummy = ltrim(dummy); // no leading spaces
+ // ignore lines with a beginning '%' or ignore all right of %
string::size_type const idx =
dummy.empty() ? string::npos : dummy.find('%');
- if (idx != string::npos) {
- // ignore lines with a beginning '%'
- if (idx > 0) {
- data_ += dummy.substr(0,data.find('%'));
- }
- } else {
- data_ += dummy;
+ if (idx != string::npos)
+ dummy.erase(idx, string::npos);
+ // do we have a new token or a new line of
+ // the same one? In the first case we ignore
+ // the \n and in the second we replace it
+ // with a space
+ if (!dummy.empty()) {
+ if (!contains(dummy, "="))
+ data_ += ' ' + dummy;
+ else
+ data_ += dummy;
}
dummy = token(data, '\n', ++Entries);
}
- data = data_;
+
+ // replace double commas with "" for easy scanning
+ data = subst(data_, ",,", "\"\"");
// unlikely!
if (data.empty())
data += ','; // now we have same behaviour for all entries
// because the last one is "blah ... }"
Entries = 0;
- dummy = token(data, ',', Entries);
- while (!contains(lowercase(dummy), findkey) && !dummy.empty())
- dummy = token(data, ',', ++Entries);
+ bool found = false;
+ // parsing of title and booktitle is different from the
+ // others, because booktitle contains title
+ do {
+ dummy = token(data, ',', Entries++);
+ if (!dummy.empty()) {
+ found = contains(ascii_lowercase(dummy), findkey);
+ if (findkey == "title" &&
+ contains(ascii_lowercase(dummy), "booktitle"))
+ found = false;
+ }
+ } while (!found && !dummy.empty());
if (dummy.empty())
- return string(); // no such keyword
- // we are not sure, if we get all, because "key= "blah, blah" is allowed.
- // therefore we read all until the next "=" character, which follows a
+ // no such keyword
+ return string();
+
+ // we are not sure, if we get all, because "key= "blah, blah" is
+ // allowed.
+ // Therefore we read all until the next "=" character, which follows a
// new keyword
keyvalue = dummy;
- dummy = token(data, ',', ++Entries);
+ dummy = token(data, ',', Entries++);
while (!contains(dummy, '=') && !dummy.empty()) {
- keyvalue += (',' + dummy);
- dummy = token(data, ',', ++Entries);
+ keyvalue += ',' + dummy;
+ dummy = token(data, ',', Entries++);
}
- data = keyvalue; // now we have the important line
- data = strip(data, ' '); // all spaces
- if (!contains(data, '{')) // no opening '{'
- data = strip(data, '}');// maybe there is a main closing '}'
+
+ // replace double "" with originals ,, (two commas)
+ // leaving us with the all-important line
+ data = subst(keyvalue, "\"\"", ",,");
+
+ // Clean-up.
+ // 1. Spaces
+ data = rtrim(data);
+ // 2. if there is no opening '{' then a closing '{' is probably cruft.
+ if (!contains(data, '{'))
+ data = rtrim(data, "}");
// happens, when last keyword
string::size_type const idx =
!data.empty() ? data.find('=') : string::npos;
if (idx == string::npos)
return string();
- data = data.substr(idx);
- data = frontStrip(strip(data));
+ data = trim(data.substr(idx));
if (data.length() < 2 || data[0] != '=') { // a valid entry?
return string();
} else {
// delete '=' and the following spaces
- data = frontStrip(frontStrip(data,'='));
+ data = ltrim(data, " =");
if (data.length() < 2) {
return data; // not long enough to find delimiters
} else {
} else {
// no {} and no "", pure data but with a
// possible ',' at the end
- return strip(data,',');
+ return rtrim(data, ",");
}
string tmp = data.substr(keypos);
while (tmp.find('{') != string::npos &&
}
+namespace {
+
+using namespace biblio;
+
+char const * const citeCommands[] = {
+ "cite", "citet", "citep", "citealt", "citealp", "citeauthor",
+ "citeyear", "citeyearpar" };
+
+unsigned int const nCiteCommands =
+ sizeof(citeCommands) / sizeof(char *);
+
+CiteStyle const citeStyles[] = {
+ CITE, CITET, CITEP, CITEALT, CITEALP,
+ CITEAUTHOR, CITEYEAR, CITEYEARPAR };
+
+unsigned int const nCiteStyles =
+ sizeof(citeStyles) / sizeof(CiteStyle);
+
+CiteStyle const citeStylesFull[] = {
+ CITET, CITEP, CITEALT, CITEALP, CITEAUTHOR };
+
+unsigned int const nCiteStylesFull =
+ sizeof(citeStylesFull) / sizeof(CiteStyle);
+
+CiteStyle const citeStylesUCase[] = {
+ CITET, CITEP, CITEALT, CITEALP, CITEAUTHOR };
+
+unsigned int const nCiteStylesUCase =
+ sizeof(citeStylesUCase) / sizeof(CiteStyle);
+
+} // namespace anon
+
+
CitationStyle const getCitationStyle(string const & command)
{
if (command.empty()) return CitationStyle();
cmd[0] = 'c';
}
- size_t n = cmd.size()-1;
+ size_t n = cmd.size() - 1;
if (cmd[n] == '*') {
cs.full = true;
cmd = cmd.substr(0,n);
if (full) {
CiteStyle const * last = citeStylesFull + nCiteStylesFull;
if (std::find(citeStylesFull, last, command) != last)
- cite += "*";
+ cite += '*';
}
if (forceUCase) {
InfoMap const & map, vector<CiteStyle> const & styles)
{
if (map.empty()) {
- vector<string> vec(1);
- vec[0] = _("No database");
- return vec;
+ return vector<string>();
}
- vector<string> vec(styles.size());
-
string const author = getAbbreviatedAuthor(map, key);
string const year = getYear(map, key);
+ if (author.empty() || year.empty())
+ return vector<string>();
+ vector<string> vec(styles.size());
for (vector<string>::size_type i = 0; i != vec.size(); ++i) {
string str;
break;
case CITEYEARPAR:
- str = "(" + year + ")";
+ str = '(' + year + ')';
break;
}
InfoMap const & map, vector<CiteStyle> const & styles)
{
if (map.empty()) {
- vector<string> vec(1);
- vec[0] = _("No database");
- return vec;
+ return vector<string>();
}
- vector<string> vec(styles.size());
-
string const author = getAbbreviatedAuthor(map, key);
string const year = getYear(map, key);
+ if (author.empty() || year.empty())
+ return vector<string>();
+ vector<string> vec(styles.size());
for (vector<string>::size_type i = 0; i != vec.size(); ++i) {
string str;
switch (styles[i]) {
+ case CITE:
case CITET:
- str = author + " (" + year + ")";
+ str = author + " (" + year + ')';
break;
- case CITE:
case CITEP:
- str = "(" + author + ", " + year + ")";
+ str = '(' + author + ", " + year + ')';
break;
case CITEALT:
- str = author + " " + year ;
+ str = author + ' ' + year ;
break;
case CITEALP:
break;
case CITEYEARPAR:
- str = "(" + year + ")";
+ str = '(' + year + ')';
break;
}