Andre Poenitz wrote:
On Tue, Jul 31, 2007 at 03:00:34PM -0400, Richard Heck wrote:
Index: src/Buffer.h
===================================================================
--- src/Buffer.h (revision 19264)
+++ src/Buffer.h (working copy)
@@ -13,7 +13,11 @@
#define BUFFER_H
#include "DocIterator.h"
+#include "ErrorList.h"
+#include "InsetList.h"
+#include "frontends/controllers/frontend_helpers.h"
Why that? Including stuff from frontends/ in src/ is wrong in theory.
Yes, I agree. This is needed, IIRC (I did this a while ago), for certain
constants, e.g.:
static const docstring
TheBibliographyRef(from_ascii("TheBibliographyRef"));
which was in frontends/controllers/frontend_helpers.cpp and which I've
moved (in effect) to the header.
There is a lot of BibTeX-related stuff in frontends/controllers/. I
think most of that should be elsewhere. I think the whole biblio
namespace should be moved, probably into src/Biblio.{h,cpp}, or
something similar. But I guess I thought I should address this first
within the existing structure and save the clean-up for later. If you
think I should just do that now, I can do that.
[...]
@@ -1359,11 +1359,12 @@
static_cast<InsetBibitem const &>(*it);
// FIXME UNICODE
string const key = to_utf8(inset.getParam("key"));
- docstring const label = inset.getParam("label");
+ biblio::KeyValMap keyvalmap;
+ keyvalmap[from_ascii("label")] =
inset.getParam("label");
DocIterator doc_it(it); doc_it.forwardPos();
- docstring const ref =
doc_it.paragraph().asString(*this, false);
- docstring const info = label + "TheBibliographyRef" +
ref;
- keys.push_back(pair<string, docstring>(key, info));
+ keyvalmap [from_ascii("ref")] =
doc_it.paragraph().asString(*this, false);
No space before [
Thanks.
[...]
-docstring const parseBibTeX(docstring data, std::string const & findkey);
+docstring const getValueForKey(KeyValMap data, std::string const & findkey);
Passing a std::map<docstring, docstring> by value seems wrong.
Yes. Made it const &. This required a const_cast, but that seems ok.
void fillWithBibKeys(Buffer const & buffer,
- std::vector<std::pair<std::string, docstring> > & keys) const;
+ std::map<std::string, std::map<docstring, docstring> > & keys)
const;
Maybe a typedef would make this clearer, too.
Done.
Thanks again. Glad someone reads these patches....
Updated patch attached.
Richard
--
==================================================================
Richard G Heck, Jr
Professor of Philosophy
Brown University
http://frege.brown.edu/heck/
==================================================================
Get my public key from http://sks.keyserver.penguin.de
Hash: 0x1DE91F1E66FFBDEC
Learn how to sign your email using Thunderbird and GnuPG at:
http://dudu.dyn.2-h.org/nist/gpg-enigmail-howto
Index: src/Buffer.h
===================================================================
--- src/Buffer.h (revision 19264)
+++ src/Buffer.h (working copy)
@@ -13,7 +13,11 @@
#define BUFFER_H
#include "DocIterator.h"
+#include "ErrorList.h"
+#include "InsetList.h"
+#include "frontends/controllers/frontend_helpers.h"
+
#include "support/FileName.h"
#include "support/limited_stack.h"
#include "support/types.h"
@@ -283,7 +287,7 @@
void validate(LaTeXFeatures &) const;
/// return all bibkeys from buffer and its childs
- void fillWithBibKeys(std::vector<std::pair<std::string, docstring> > & keys) const;
+ void fillWithBibKeys(biblio::BibKeyList & keys) const;
/// Update the cache with all bibfiles in use (including bibfiles
/// of loaded child documents).
void updateBibfilesCache();
Index: src/Buffer.cpp
===================================================================
--- src/Buffer.cpp (revision 19264)
+++ src/Buffer.cpp (working copy)
@@ -1333,7 +1333,7 @@
// This is also a buffer property (ale)
-void Buffer::fillWithBibKeys(vector<pair<string, docstring> > & keys)
+void Buffer::fillWithBibKeys(biblio::BibKeyList & keys)
const
{
/// if this is a child document and the parent is already loaded
@@ -1359,11 +1359,12 @@
static_cast<InsetBibitem const &>(*it);
// FIXME UNICODE
string const key = to_utf8(inset.getParam("key"));
- docstring const label = inset.getParam("label");
+ biblio::KeyValMap keyvalmap;
+ keyvalmap[from_ascii("label")] = inset.getParam("label");
DocIterator doc_it(it); doc_it.forwardPos();
- docstring const ref = doc_it.paragraph().asString(*this, false);
- docstring const info = label + "TheBibliographyRef" + ref;
- keys.push_back(pair<string, docstring>(key, info));
+ keyvalmap[from_ascii("ref")] = doc_it.paragraph().asString(*this, false);
+ keyvalmap[biblio::TheBibliographyRef] = biblio::TheBibliographyRef;
+ keys[key] = keyvalmap;
}
}
}
@@ -1725,10 +1726,10 @@
vector<docstring> labels;
if (code == Inset::CITE_CODE) {
- vector<pair<string, docstring> > keys;
+ biblio::BibKeyList keys;
fillWithBibKeys(keys);
- vector<pair<string, docstring> >::const_iterator bit = keys.begin();
- vector<pair<string, docstring> >::const_iterator bend = keys.end();
+ biblio::BibKeyList::const_iterator bit = keys.begin();
+ biblio::BibKeyList::const_iterator bend = keys.end();
for (; bit != bend; ++bit)
// FIXME UNICODE
Index: src/frontends/controllers/frontend_helpers.h
===================================================================
--- src/frontends/controllers/frontend_helpers.h (revision 19264)
+++ src/frontends/controllers/frontend_helpers.h (working copy)
@@ -28,8 +28,18 @@
/** Functions of use to citation and bibtex GUI controllers and views */
namespace lyx {
+
namespace biblio {
+
+/// First entry is field, second is value
+typedef std::map<docstring, docstring> KeyValMap;
+/// First entry is the bibliography key, second the data
+typedef std::map<std::string, std::map<docstring, docstring> > BibKeyList;
+static const docstring TheBibliographyRef(from_ascii("@LyXInfo"));
+static const docstring TheDataString(from_ascii("@BibTeXData"));
+static const docstring TheEntryType(from_ascii("@BibTeXEntryType"));
+
enum CiteEngine {
ENGINE_BASIC,
ENGINE_NATBIB_AUTHORYEAR,
@@ -69,34 +79,31 @@
std::string const asValidLatexCommand(std::string const & input,
CiteEngine const engine);
-/// First entry is the bibliography key, second the data
-typedef std::map<std::string, docstring> InfoMap;
-
/// Returns a vector of bibliography keys
-std::vector<std::string> const getKeys(InfoMap const &);
+std::vector<std::string> const getKeys(BibKeyList const &);
/** Returns the BibTeX data associated with a given key.
- Empty if no info exists. */
-docstring const getInfo(InfoMap const &, std::string const & key);
+ Empty if the key was not defined in the BibTeX record. */
+docstring const getInfo(BibKeyList const &, std::string const & key);
/// return the year from the bibtex data record
-docstring const getYear(InfoMap const & map, std::string const & key);
+docstring const getYear(BibKeyList const & map, std::string const & key);
/// return the short form of an authorlist
-docstring const getAbbreviatedAuthor(InfoMap const & map, std::string const & key);
+docstring const getAbbreviatedAuthor(BibKeyList const & map, std::string const & key);
// return only the family name
docstring const familyName(docstring const & name);
/** Search a BibTeX info field for the given key and return the
associated field. */
-docstring const parseBibTeX(docstring data, std::string const & findkey);
+docstring const getValueForKey(KeyValMap const & data, std::string const & findkey);
/** Returns an iterator to the first key that meets the search
criterion, or end() if unsuccessful.
User supplies :
- the InfoMap of bibkeys info,
+ the BibKeyList of bibkeys info,
the vector of keys to be searched,
the search criterion,
an iterator defining the starting point of the search,
@@ -105,7 +112,7 @@
*/
std::vector<std::string>::const_iterator
-searchKeys(InfoMap const & map,
+searchKeys(BibKeyList const & map,
std::vector<std::string> const & keys_to_search,
docstring const & search_expression,
std::vector<std::string>::const_iterator start,
@@ -145,12 +152,12 @@
User supplies :
the key,
- the InfoMap of bibkeys info,
+ the BibKeyList of bibkeys info,
the available citation styles
*/
std::vector<docstring> const
getNumericalStrings(std::string const & key,
- InfoMap const & map,
+ BibKeyList const & map,
std::vector<CiteStyle> const & styles);
/**
@@ -162,12 +169,12 @@
User supplies :
the key,
- the InfoMap of bibkeys info,
+ the BibKeyList of bibkeys info,
the available citation styles
*/
std::vector<docstring> const
getAuthorYearStrings(std::string const & key,
- InfoMap const & map,
+ BibKeyList const & map,
std::vector<CiteStyle> const & styles);
} // namespace biblio
Index: src/frontends/controllers/frontend_helpers.cpp
===================================================================
--- src/frontends/controllers/frontend_helpers.cpp (revision 19264)
+++ src/frontends/controllers/frontend_helpers.cpp (working copy)
@@ -5,6 +5,7 @@
*
* \author Angus Leeming
* \author Herbert Voß
+ * \author Richard Heck (improvements of BibTeX stuff)
*
* Full author contact details are available in file CREDITS.
*/
@@ -26,9 +27,6 @@
#include "support/filetools.h"
#include "support/lstrings.h"
-#include "support/Package.h"
-#include "support/filetools.h"
-#include "support/lstrings.h"
#include "support/lyxalgo.h"
#include "support/os.h"
#include "support/Package.h"
@@ -123,9 +121,6 @@
return str;
}
-
-static const docstring TheBibliographyRef(from_ascii("TheBibliographyRef"));
-
} // namespace anon
@@ -200,38 +195,29 @@
}
-docstring const getAbbreviatedAuthor(InfoMap const & map, string const & key)
+docstring const getAbbreviatedAuthor(BibKeyList const & map, string const & key)
{
BOOST_ASSERT(!map.empty());
- InfoMap::const_iterator it = map.find(key);
+ BibKeyList::const_iterator it = map.find(key);
if (it == map.end())
return docstring();
- docstring const & data = it->second;
+ KeyValMap const & data = it->second;
// Is the entry a BibTeX one or one from lyx-layout "bibliography"?
- docstring::size_type const pos = data.find(TheBibliographyRef);
- if (pos != docstring::npos) {
- if (pos <= 2) {
- return docstring();
- }
+ KeyValMap::const_iterator it2 = data.find(TheBibliographyRef);
+ if (it2 != data.end())
+ // We don't have any way to tell how the author names might have
+ // been formatted.
+ return docstring();
- docstring const opt = trim(data.substr(0, pos - 1));
- if (opt.empty())
- return docstring();
+ docstring author = getValueForKey(data, "author");
- docstring authors;
- split(opt, authors, '(');
- return authors;
- }
-
- docstring author = parseBibTeX(data, "author");
-
if (author.empty())
- author = parseBibTeX(data, "editor");
+ author = getValueForKey(data, "editor");
if (author.empty()) {
- author = parseBibTeX(data, "key");
+ author = getValueForKey(data, "key");
if (author.empty())
// FIXME UNICODE
return from_utf8(key);
@@ -253,36 +239,23 @@
}
-docstring const getYear(InfoMap const & map, string const & key)
+docstring const getYear(BibKeyList const & map, string const & key)
{
BOOST_ASSERT(!map.empty());
- InfoMap::const_iterator it = map.find(key);
+ BibKeyList::const_iterator it = map.find(key);
if (it == map.end())
return docstring();
- docstring const & data = it->second;
+ KeyValMap const & data = it->second;
// Is the entry a BibTeX one or one from lyx-layout "bibliography"?
- docstring::size_type const pos = data.find(TheBibliographyRef);
- if (pos != docstring::npos) {
- if (pos <= 2) {
- return docstring();
- }
+ KeyValMap::const_iterator it2 = data.find(TheBibliographyRef);
+ if (it2 != data.end())
+ // We don't have any way to tell how the entry might have
+ // been formatted.
+ return docstring();
- docstring const opt =
- trim(data.substr(0, pos - 1));
- if (opt.empty())
- return docstring();
-
- docstring authors;
- docstring const tmp = split(opt, authors, '(');
- docstring year;
- split(tmp, year, ')');
- return year;
-
- }
-
- docstring year = parseBibTeX(data, "year");
+ docstring year = getValueForKey(data, "year");
if (year.empty())
year = _("No year");
@@ -304,11 +277,11 @@
} // namespace anon
-vector<string> const getKeys(InfoMap const & map)
+vector<string> const getKeys(BibKeyList const & map)
{
vector<string> bibkeys;
- InfoMap::const_iterator it = map.begin();
- InfoMap::const_iterator end = map.end();
+ BibKeyList::const_iterator it = map.begin();
+ BibKeyList::const_iterator end = map.end();
for (; it != end; ++it) {
bibkeys.push_back(it->first);
}
@@ -318,72 +291,69 @@
}
-docstring const getInfo(InfoMap const & map, string const & key)
+docstring const getInfo(BibKeyList const & map, string const & key)
{
BOOST_ASSERT(!map.empty());
- InfoMap::const_iterator it = map.find(key);
+ BibKeyList::const_iterator it = map.find(key);
if (it == map.end())
return docstring();
- docstring const & data = it->second;
+ KeyValMap const & data = it->second;
- // is the entry a BibTeX one or one from lyx-layout "bibliography"?
- docstring::size_type const pos = data.find(TheBibliographyRef);
- if (pos != docstring::npos) {
- docstring::size_type const pos2 = pos + TheBibliographyRef.size();
- docstring const info = trim(data.substr(pos2));
- return info;
+ // Is the entry a BibTeX one or one from lyx-layout "bibliography"?
+ KeyValMap::const_iterator it2 = data.find(TheBibliographyRef);
+ if (it2 != data.end()) {
+ KeyValMap::const_iterator it3 = data.find(from_ascii("ref"));
+ return it3->second;
}
- // Search for all possible "required" keys
- docstring author = parseBibTeX(data, "author");
+ //FIXME FIXME FIXME
+ //This could be made alot better using the biblio::TheEntryType
+ //field to customize the output based upon entry type.
+
+ //Search for all possible "required" fields
+ docstring author = getValueForKey(data, "author");
if (author.empty())
- author = parseBibTeX(data, "editor");
+ author = getValueForKey(data, "editor");
- docstring year = parseBibTeX(data, "year");
- docstring title = parseBibTeX(data, "title");
- docstring booktitle = parseBibTeX(data, "booktitle");
- docstring chapter = parseBibTeX(data, "chapter");
- docstring number = parseBibTeX(data, "number");
- docstring volume = parseBibTeX(data, "volume");
- docstring pages = parseBibTeX(data, "pages");
- docstring annote = parseBibTeX(data, "annote");
- docstring media = parseBibTeX(data, "journal");
- if (media.empty())
- media = parseBibTeX(data, "publisher");
- if (media.empty())
- media = parseBibTeX(data, "school");
- if (media.empty())
- media = parseBibTeX(data, "institution");
+ docstring year = getValueForKey(data, "year");
+ docstring title = getValueForKey(data, "title");
+ docstring docLoc = getValueForKey(data, "pages");
+ if (docLoc.empty()) {
+ docLoc = getValueForKey(data, "chapter");
+ if (!docLoc.empty())
+ docLoc = from_ascii("Ch. ") + docLoc;
+ } else
+ docLoc = from_ascii("pp. ") + docLoc;
+ docstring media = getValueForKey(data, "journal");
+ if (media.empty()) {
+ media = getValueForKey(data, "publisher");
+ if (media.empty()) {
+ media = getValueForKey(data, "school");
+ if (media.empty())
+ media = getValueForKey(data, "institution");
+ }
+ }
+ docstring volume = getValueForKey(data, "volume");
odocstringstream result;
if (!author.empty())
result << author << ", ";
if (!title.empty())
result << title;
- if (!booktitle.empty())
- result << ", in " << booktitle;
- if (!chapter.empty())
- result << ", Ch. " << chapter;
if (!media.empty())
result << ", " << media;
- if (!volume.empty())
- result << ", vol. " << volume;
- if (!number.empty())
- result << ", no. " << number;
- if (!pages.empty())
- result << ", pp. " << pages;
if (!year.empty())
result << ", " << year;
- if (!annote.empty())
- result << "\n\n" << annote;
+ if (!docLoc.empty())
+ result << ", " << docLoc;
docstring const result_str = rtrim(result.str());
if (!result_str.empty())
return result_str;
// This should never happen (or at least be very unusual!)
- return data;
+ return docstring();
}
@@ -420,24 +390,28 @@
public:
// 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)
+ RegexMatch(BibKeyList const & m, string const & re, bool icase)
: map_(m), regex_(re, icase) {}
+ //FIXME This should probably be restored to its earlier behavior
bool operator()(string const & key) const {
- // 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())
+ BibKeyList::const_iterator info = map_.find(key);
+ if (info == map_.end())
+ return false;
+
+ string data = key;
+ //The machinations here are required because map::operator[]
+ //has no const version.
+ KeyValMap const kvm = info->second;
+ KeyValMap::const_iterator it = kvm.find(TheDataString);
+ if (it != kvm.end())
// FIXME UNICODE
- data += ' ' + to_utf8(info->second);
-
- // Attempts to find a match for the current RE
- // somewhere in data.
+ data += ' ' + to_utf8(it->second);
+
return boost::regex_search(data, regex_);
}
private:
- InfoMap const map_;
+ BibKeyList const map_;
mutable boost::regex regex_;
};
@@ -445,7 +419,7 @@
vector<string>::const_iterator
-searchKeys(InfoMap const & theMap,
+searchKeys(BibKeyList const & theMap,
vector<string> const & keys,
string const & search_expr,
vector<string>::const_iterator start,
@@ -492,142 +466,14 @@
}
-docstring const parseBibTeX(docstring data, string const & findkey)
+docstring const getValueForKey(KeyValMap const & data, string const & findkey)
{
- // 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.
- docstring data_;
- int Entries = 0;
- docstring dummy = token(data,'\n', Entries);
- while (!dummy.empty()) {
- // no tabs
- dummy = subst(dummy, '\t', ' ');
- // no leading spaces
- dummy = ltrim(dummy);
- // ignore lines with a beginning '%' or ignore all right of %
- docstring::size_type const idx =
- dummy.empty() ? docstring::npos : dummy.find('%');
- if (idx != docstring::npos)
- // Check if this is really a comment or just "\%"
- if (idx == 0 || dummy[idx - 1] != '\\')
- dummy.erase(idx, docstring::npos);
- else
- // This is "\%", so just erase the '\'
- dummy.erase(idx - 1, 1);
- // 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);
- }
-
- // replace double commas with "" for easy scanning
- data = subst(data_, from_ascii(",,"), from_ascii("\"\""));
-
- // unlikely!
- if (data.empty())
+ docstring key = from_ascii(findkey);
+ KeyValMap::const_iterator it = data.find(key);
+ if (it == data.end())
return docstring();
-
- // now get only the important line of the bibtex entry.
- // all entries are devided by ',' except the last one.
- data += ',';
- // now we have same behaviour for all entries because the last one
- // is "blah ... }"
- Entries = 0;
- 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), from_ascii(findkey));
- if (findkey == "title" &&
- contains(ascii_lowercase(dummy), from_ascii("booktitle")))
- found = false;
- }
- } while (!found && !dummy.empty());
- if (dummy.empty())
- // no such keyword
- return docstring();
-
- // 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
- docstring keyvalue = dummy;
- dummy = token(data, ',', Entries++);
- while (!contains(dummy, '=') && !dummy.empty()) {
- keyvalue += ',' + dummy;
- dummy = token(data, ',', Entries++);
- }
-
- // replace double "" with originals ,, (two commas)
- // leaving us with the all-important line
- data = subst(keyvalue, from_ascii("\"\""), from_ascii(",,"));
-
- // 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
- docstring::size_type const idx =
- !data.empty() ? data.find('=') : docstring::npos;
-
- if (idx == docstring::npos)
- return docstring();
-
- data = trim(data.substr(idx));
-
- // a valid entry?
- if (data.length() < 2 || data[0] != '=')
- return docstring();
- else {
- // delete '=' and the following spaces
- data = ltrim(data, " =");
- if (data.length() < 2) {
- // not long enough to find delimiters
- return data;
- } else {
- docstring::size_type keypos = 1;
- char_type enclosing;
- if (data[0] == '{') {
- enclosing = '}';
- } else if (data[0] == '"') {
- enclosing = '"';
- } else {
- // no {} and no "", pure data but with a
- // possible ',' at the end
- return rtrim(data, ",");
- }
- docstring tmp = data.substr(keypos);
- while (tmp.find('{') != docstring::npos &&
- tmp.find('}') != docstring::npos &&
- tmp.find('{') < tmp.find('}') &&
- tmp.find('{') < tmp.find(enclosing)) {
-
- keypos += tmp.find('{') + 1;
- tmp = data.substr(keypos);
- keypos += tmp.find('}') + 1;
- tmp = data.substr(keypos);
- }
- if (tmp.find(enclosing) == docstring::npos)
- return data;
- else {
- keypos += tmp.find(enclosing);
- return data.substr(1, keypos - 1);
- }
- }
- }
+ KeyValMap & data2 = const_cast<KeyValMap &>(data);
+ return data2[key];
}
@@ -745,7 +591,7 @@
vector<docstring> const
getNumericalStrings(string const & key,
- InfoMap const & map, vector<CiteStyle> const & styles)
+ BibKeyList const & map, vector<CiteStyle> const & styles)
{
if (map.empty())
return vector<docstring>();
@@ -799,7 +645,7 @@
vector<docstring> const
getAuthorYearStrings(string const & key,
- InfoMap const & map, vector<CiteStyle> const & styles)
+ BibKeyList const & map, vector<CiteStyle> const & styles)
{
if (map.empty())
return vector<docstring>();
Index: src/frontends/controllers/ControlCitation.h
===================================================================
--- src/frontends/controllers/ControlCitation.h (revision 19264)
+++ src/frontends/controllers/ControlCitation.h (working copy)
@@ -63,7 +63,7 @@
}
private:
/// The info associated with each key
- biblio::InfoMap bibkeysInfo_;
+ biblio::BibKeyList bibkeysInfo_;
///
static std::vector<biblio::CiteStyle> citeStyles_;
Index: src/frontends/controllers/ControlCitation.cpp
===================================================================
--- src/frontends/controllers/ControlCitation.cpp (revision 19264)
+++ src/frontends/controllers/ControlCitation.cpp (working copy)
@@ -12,6 +12,7 @@
#include <config.h>
#include "ControlCitation.h"
+#include "frontend_helpers.h"
#include "Buffer.h"
#include "BufferParams.h"
@@ -48,12 +49,8 @@
bool use_styles = engine != biblio::ENGINE_BASIC;
- vector<pair<string, docstring> > blist;
- kernel().buffer().fillWithBibKeys(blist);
- bibkeysInfo_.clear();
- for (size_t i = 0; i < blist.size(); ++i)
- bibkeysInfo_[blist[i].first] = blist[i].second;
-
+ kernel().buffer().fillWithBibKeys(bibkeysInfo_);
+
if (citeStyles_.empty())
citeStyles_ = biblio::getCiteStyles(engine);
else {
@@ -137,23 +134,21 @@
// it is treated as a simple string by boost::regex.
expr = escape_special_chars(expr);
- boost::regex reg_exp(to_utf8(expr), case_sensitive?
+ boost::regex reg_exp(to_utf8(expr), case_sensitive ?
boost::regex_constants::normal : boost::regex_constants::icase);
vector<string>::const_iterator it = keys_to_search.begin();
vector<string>::const_iterator end = keys_to_search.end();
for (; it != end; ++it ) {
- biblio::InfoMap::const_iterator info = bibkeysInfo_.find(*it);
+ biblio::BibKeyList::iterator info = bibkeysInfo_.find(*it);
if (info == bibkeysInfo_.end())
continue;
string data = *it;
// FIXME UNICODE
- data += ' ' + to_utf8(info->second);
+ data += ' ' + to_utf8((info->second)[biblio::TheDataString]);
try {
- // Attempts to find a match for the current RE
- // somewhere in data.
if (boost::regex_search(data, reg_exp))
foundKeys.push_back(*it);
}
Index: src/insets/InsetBibtex.h
===================================================================
--- src/insets/InsetBibtex.h (revision 19264)
+++ src/insets/InsetBibtex.h (working copy)
@@ -13,8 +13,9 @@
#define INSET_BIBTEX_H
-#include <vector>
+#include <map>
#include "InsetCommand.h"
+#include "frontends/controllers/frontend_helpers.h"
#include "support/FileName.h"
@@ -38,8 +39,7 @@
///
int latex(Buffer const &, odocstream &, OutputParams const &) const;
///
- void fillWithBibKeys(Buffer const & buffer,
- std::vector<std::pair<std::string, docstring> > & keys) const;
+ void fillWithBibKeys(Buffer const & buffer, biblio::BibKeyList & keys) const;
///
std::vector<support::FileName> const getFiles(Buffer const &) const;
///
Index: src/insets/InsetBibtex.cpp
===================================================================
--- src/insets/InsetBibtex.cpp (revision 19264)
+++ src/insets/InsetBibtex.cpp (working copy)
@@ -4,6 +4,7 @@
* Licence details can be found in the file COPYING.
*
* \author Alejandro Aguilar Sierra
+ * \author Richard Heck (BibTeX parser improvements)
*
* Full author contact details are available in file CREDITS.
*/
@@ -34,7 +35,6 @@
#include <boost/tokenizer.hpp>
-
namespace lyx {
using support::absolutePath;
@@ -415,13 +415,13 @@
bool legalChar = true;
while (ifs && !isSpace(ch) &&
delimChars.find(ch) == docstring::npos &&
- (legalChar = illegalChars.find(ch) == docstring::npos)
- ) {
- if (chCase == makeLowerCase) {
+ (legalChar = (illegalChars.find(ch) == docstring::npos))
+ )
+ {
+ if (chCase == makeLowerCase)
val += lowercase(ch);
- } else {
+ else
val += ch;
- }
ifs.get(ch);
}
@@ -478,19 +478,41 @@
return false;
} else if (ch == '"' || ch == '{') {
+ // set end delimiter
+ char_type delim = ch == '"' ? '"': '}';
- // read delimited text - set end delimiter
- char_type delim = ch == '"'? '"': '}';
-
+ //Skip whitespace
+ do {
+ ifs.get(ch);
+ } while (ifs && isSpace(ch));
+
+ if (!ifs)
+ return false;
+
+ //We now have the first non-whitespace character
+ //We'll collapse adjacent whitespace.
+ bool lastWasWhiteSpace = false;
+
// inside this delimited text braces must match.
// Thus we can have a closing delimiter only
// when nestLevel == 0
int nestLevel = 0;
- ifs.get(ch);
while (ifs && (nestLevel > 0 || ch != delim)) {
+ if (isSpace(ch)) {
+ lastWasWhiteSpace = true;
+ ifs.get(ch);
+ continue;
+ }
+ //We output the space only after we stop getting
+ //whitespace so as not to output any whitespace
+ //at the end of the value.
+ if (lastWasWhiteSpace) {
+ lastWasWhiteSpace = false;
+ val += ' ';
+ }
+
val += ch;
-
// update nesting level
switch (ch) {
case '{':
@@ -503,7 +525,7 @@
}
ifs.get(ch);
- }
+ } //end while loop
if (!ifs)
return false;
@@ -554,9 +576,9 @@
}
-// This method returns a comma separated list of Bibtex entries
+// This method returns a map of Bibtex entries
void InsetBibtex::fillWithBibKeys(Buffer const & buffer,
- std::vector<std::pair<string, docstring> > & keys) const
+ biblio::BibKeyList & keys) const
{
vector<FileName> const files = getFiles(buffer);
for (vector<FileName>::const_iterator it = files.begin();
@@ -571,15 +593,6 @@
// field values.
// - it accepts more characters in keys or value names than
// bibtex does.
- //
- // TODOS:
- // - the entries are split into name = value pairs by the
- // parser. These have to be merged again because of the
- // way lyx treats the entries ( pair<...>(...) ). The citation
- // mechanism in lyx should be changed such that it can use
- // the split entries.
- // - messages on parsing errors can be generated.
- //
// Officially bibtex does only support ASCII, but in practice
// you can use the encoding of the main document as long as
@@ -588,6 +601,7 @@
// We don't restrict keys to ASCII in LyX, since our own
// InsetBibitem can generate non-ASCII keys, and nonstandard
// 8bit clean bibtex forks exist.
+
idocfstream ifs(it->toFilesystemEncoding().c_str(),
std::ios_base::in,
buffer.params().encoding().iconvName());
@@ -658,24 +672,29 @@
continue;
} else {
-
- // Citation entry. Read the key and all name = value pairs
+ // Citation entry. Try to read the key.
docstring key;
- docstring fields;
- docstring name;
- docstring value;
- docstring commaNewline;
if (!readTypeOrKey(key, ifs, from_ascii(","),
from_ascii("}"), keepCase) || !ifs)
continue;
- // now we have a key, so we will add an entry
+ /////////////////////////////////////////////
+ // now we have a key, so we will add an entry
// (even if it's empty, as bibtex does)
//
+ // we now read the field = value pairs.
// all items must be separated by a comma. If
// it is missing the scanning of this entry is
// stopped and the next is searched.
+ docstring fields;
+ docstring name;
+ docstring value;
+ docstring commaNewline;
+ docstring data;
+ biblio::KeyValMap keyvalmap;
+ keyvalmap[biblio::TheEntryType] = entryType;
+
bool readNext = removeWSAndComma(ifs);
while (ifs && readNext) {
@@ -698,23 +717,15 @@
if (!readValue(value, ifs, strings))
break;
- // append field to the total entry string.
- //
- // TODO: Here is where the fields can be put in
- // a more intelligent structure that preserves
- // the already known parts.
- fields += commaNewline;
- fields += name + from_ascii(" = {") + value + '}';
-
- if (!commaNewline.length())
- commaNewline = from_ascii(",\n");
-
+ keyvalmap[name] = value;
+ data += "\n\n" + value;
+
readNext = removeWSAndComma(ifs);
}
// add the new entry
- keys.push_back(pair<string, docstring>(
- to_utf8(key), fields));
+ keyvalmap[biblio::TheDataString] = data;
+ keys[to_utf8(key)] = keyvalmap;
}
} //< searching '@'
Index: src/insets/InsetCitation.cpp
===================================================================
--- src/insets/InsetCitation.cpp (revision 19264)
+++ src/insets/InsetCitation.cpp (working copy)
@@ -65,13 +65,13 @@
return docstring();
// Cache the labels
- typedef std::map<Buffer const *, biblio::InfoMap> CachedMap;
+ typedef std::map<Buffer const *, biblio::BibKeyList> CachedMap;
static CachedMap cached_keys;
// and cache the timestamp of the bibliography files.
static std::map<FileName, time_t> bibfileStatus;
- biblio::InfoMap infomap;
+ biblio::BibKeyList infomap;
vector<FileName> const & bibfilesCache = buffer.getBibfilesCache();
// compare the cached timestamps with the actual ones.
@@ -97,16 +97,7 @@
// build the keylist only if the bibfiles have been changed
if (cached_keys[&buffer].empty() || bibfileStatus.empty() || changed) {
- typedef vector<std::pair<string, docstring> > InfoType;
- InfoType bibkeys;
- buffer.fillWithBibKeys(bibkeys);
-
- InfoType::const_iterator bit = bibkeys.begin();
- InfoType::const_iterator bend = bibkeys.end();
-
- for (; bit != bend; ++bit)
- infomap[bit->first] = bit->second;
-
+ buffer.fillWithBibKeys(infomap);
cached_keys[&buffer] = infomap;
} else
// use the cached keys
Index: src/insets/InsetInclude.h
===================================================================
--- src/insets/InsetInclude.h (revision 19264)
+++ src/insets/InsetInclude.h (working copy)
@@ -18,6 +18,8 @@
#include "MailInset.h"
#include "Counters.h"
+#include "frontends/controllers/frontend_helpers.h"
+
#include "support/FileName.h"
#include <boost/scoped_ptr.hpp>
@@ -57,10 +59,9 @@
std::vector<docstring> & list) const;
/** Fills \c keys
* \param buffer the Buffer containing this inset.
- * \param keys the list of bibkeys in the child buffer.
- */
- void fillWithBibKeys(Buffer const & buffer,
- std::vector<std::pair<std::string, docstring> > & keys) const;
+ * \param keys the map of bibkeys in the child buffer.
+ */
+ void fillWithBibKeys(Buffer const & buffer, biblio::BibKeyList & keys) const;
/** Update the cache with all bibfiles in use of the child buffer
* (including bibfiles of grandchild documents).
* Does nothing if the child document is not loaded to prevent
Index: src/insets/InsetInclude.cpp
===================================================================
--- src/insets/InsetInclude.cpp (revision 19264)
+++ src/insets/InsetInclude.cpp (working copy)
@@ -736,7 +736,7 @@
void InsetInclude::fillWithBibKeys(Buffer const & buffer,
- std::vector<std::pair<string, docstring> > & keys) const
+ std::map<std::string, std::map<docstring, docstring> > & keys) const
{
if (loadIfNeeded(buffer, params_)) {
string const included_file = includedFilename(buffer, params_).absFilename();
