Hello,
please, find attached a preliminary patch implementing the idea.
Now, each paragraph that does not contain any match is only stringified
once, then we switch to the next one. When a matching paragraph
is found, it is searched for more carefully by going pos by pos.
Improvement w.r.t. the old unconditional pos-by-pos search is significant
(search in the math manual completes in a few seconds with debug enabled).
Forward search seems ok, backward search has problems, but I'd like to
collect preliminary comments about the approach.
Also, I had to add a ::stringify() method in Paragraph, that mimics the old
behaviour of the Paragraph::asString(), as now the latter method seems
to have been switched to a ToC-oriented usage.
Bye,
T.
Tommaso Cucinotta ha scritto:
I think I can easily modify it so as to realise a hierarchically
sequential search
(not a real binary search). Instead of advancing one position,
exporting from
the current position to the end of paragraph, then searching for a
matching segment
*at the start*, I would try to advance one paragraph each time,
exporting the
entire paragraph, searching for any match *inside*, then, if a match
is found,
proceed recursively advancing one position at a time searching for the
exact
position that has the match at beginning.
I'll see if I can play a little bit tonight with the idea.
T.
Abdelrazak Younes ha scritto:
Yes that too ;-)
That's also why the binary search is the most practical choice...
Abdel.
--
Tommaso Cucinotta, Computer Engineering PhD, Researcher
ReTiS Lab, Scuola Superiore Sant'Anna, Pisa, Italy
Tel +39 050 882 024, Fax +39 050 882 003
http://feanor.sssup.it/~tommaso
Index: src/lyxfind.cpp
===================================================================
--- src/lyxfind.cpp (revisione 27615)
+++ src/lyxfind.cpp (copia locale)
@@ -45,6 +45,7 @@
#include "support/lstrings.h"
#include <boost/regex.hpp>
+#include <boost/next_prior.hpp>
using namespace std;
using namespace lyx::support;
@@ -534,10 +535,14 @@
/** Tests if text starting at the supplied position matches with the one provided to the MatchStringAdv
** constructor as opt.search, under the opt.* options settings.
**
+ ** @param at_begin
+ ** If set, then match is searched only against beginning of text starting at cur.
+ ** If unset, then match is searched anywhere in text starting at cur.
+ **
** @return
** The length of the matching text, or zero if no match was found.
**/
- int operator()(DocIterator const & cur, int len = -1) const;
+ int operator()(DocIterator const & cur, int len = -1, bool at_begin = true) const;
public:
/// buffer
@@ -564,6 +569,8 @@
string par_as_string;
// regular expression to use for searching
boost::regex regexp;
+ // same as regexp, but prefixed with a ".*"
+ boost::regex regexp2;
// unmatched open braces in the search string/regexp
int open_braces;
// number of (.*?) subexpressions added at end of search regexp for closing
@@ -625,21 +632,30 @@
par_as_string = string("\\`") + par_as_string;
LYXERR(Debug::DEBUG, "Replaced text (to be used as regex): " << par_as_string);
regexp = boost::regex(par_as_string);
+ regexp2 = boost::regex(string(".*") + par_as_string);
}
}
-int MatchStringAdv::operator()(DocIterator const & cur, int len) const
+
+int MatchStringAdv::operator()(DocIterator const & cur, int len, bool at_begin) const
{
docstring docstr = stringifyFromForSearch(opt, buf, cur, len);
LYXERR(Debug::DEBUG, "Matching against '" << lyx::to_utf8(docstr) << "'");
string str = normalize(docstr);
LYXERR(Debug::DEBUG, "After normalization: '" << str << "'");
if (! opt.regexp) {
- if (str.substr(0, par_as_string.size()) == par_as_string)
- return par_as_string.size();
+ if (at_begin) {
+ if (str.substr(0, par_as_string.size()) == par_as_string)
+ return par_as_string.size();
+ } else {
+ size_t pos = str.find(par_as_string);
+ if (pos != string::npos)
+ return par_as_string.size();
+ }
} else {
// Try all possible regexp matches, until one that verifies the braces match test is found
- boost::sregex_iterator re_it(str.begin(), str.end(), regexp);
+ boost::regex const *p_regexp = at_begin ? ®exp : ®exp2;
+ boost::sregex_iterator re_it(str.begin(), str.end(), *p_regexp);
boost::sregex_iterator re_it_end;
for (; re_it != re_it_end; ++re_it) {
boost::match_results<string::const_iterator> const & m = *re_it;
@@ -693,7 +709,16 @@
Paragraph const & par = cur.paragraph();
// TODO what about searching beyond/across paragraph breaks ?
// TODO Try adding a AS_STR_INSERTS as last arg
- return par.asString(cur.pos(), ( len == -1 || cur.pos() + len > int(par.size()) ) ? int(par.size()) : cur.pos() + len, AS_STR_INSETS);
+ pos_type end = ( len == -1 || cur.pos() + len > int(par.size()) ) ? int(par.size()) : cur.pos() + len;
+ OutputParams runparams(&cur.buffer()->params().encoding());
+ odocstringstream os;
+ runparams.nice = true;
+ runparams.flavor = OutputParams::LATEX;
+ runparams.linelen = 100000; //lyxrc.plaintext_linelen;
+ // No side effect of file copying and image conversion
+ runparams.dryrun = true;
+ LYXERR(Debug::DEBUG, "Stringifying with cur: " << cur << ", from pos: " << cur.pos() << ", end: " << end);
+ return par.stringify(cur.pos(), end, AS_STR_INSETS, runparams);
} else if (cur.inMathed()) {
odocstringstream os;
CursorSlice cs = cur.top();
@@ -795,7 +820,7 @@
size_t d;
DocIterator old_cur(cur.buffer());
do {
- LYXERR(Debug::DEBUG, "Forwarding one step (searching for innermost match)");
+ LYXERR(Debug::DEBUG, "Forwarding one step (searching for innermost match)");
d = cur.depth();
old_cur = cur;
cur.forwardPos();
@@ -823,58 +848,93 @@
return len;
}
+
/// Finds forward
int findForwardAdv(DocIterator & cur, MatchStringAdv const & match)
{
if (!cur)
return 0;
- for (; cur; cur.forwardPos()) {
- // odocstringstream ods;
- // ods << _("Searching ... ")
- // << (cur.bottom().lastpit() - cur.bottom().pit()) * 100 / total;
- // cur.message(ods.str());
- if (match(cur))
- return findAdvFinalize(cur, match);
- }
+ int wrap_answer;
+ do {
+ while (cur && !match(cur, -1, false)) {
+ if (cur.pit() < cur.lastpit())
+ cur.forwardPar();
+ else {
+ cur.forwardPos();
+ }
+ }
+ for (; cur; cur.forwardPos()) {
+ if (match(cur))
+ return findAdvFinalize(cur, match);
+ }
+ wrap_answer = frontend::Alert::prompt(
+ _("Wrap search ?"),
+ _("End of document reached while searching forward\n"
+ "\n"
+ "Continue searching from beginning ?"),
+ 0, 1, _("&Yes"), _("&No"));
+ cur.clear();
+ cur.push_back(CursorSlice(match.buf.inset()));
+ } while (wrap_answer == 0);
return 0;
}
+
/// Finds backwards
-int findBackwardsAdv(DocIterator & cur, MatchStringAdv const & match)
-{
+int findBackwardsAdv(DocIterator & cur, MatchStringAdv const & match) {
// if (cur.pos() > 0 || cur.depth() > 0)
// cur.backwardPos();
DocIterator cur_orig(cur);
if (match(cur_orig))
findAdvFinalize(cur_orig, match);
- // int total = cur.bottom().pit() + 1;
- for (; cur; cur.backwardPos()) {
- // odocstringstream ods;
- // ods << _("Searching ... ") << (total - cur.bottom().pit()) * 100 / total;
- // cur.message(ods.str());
- if (match(cur)) {
- // Find the most backward consecutive match within same
- // paragraph while searching backwards.
- int pit = cur.pit();
- int old_len;
- DocIterator old_cur;
- int len = findAdvFinalize(cur, match);
- do {
- old_cur = cur;
- old_len = len;
+ // int total = cur.bottom().pit() + 1;
+ int wrap_answer;
+ do {
+ // TODO No ! così non va.
+ bool pit_changed = false;
+ while (cur && !match(cur, -1, false)) {
+ if (cur.pit() > 0)
+ --cur.pit();
+ else {
cur.backwardPos();
- LYXERR(Debug::DEBUG, "old_cur: " << old_cur
- << ", old_len: " << len << ", cur: " << cur);
- } while (cur && cur.pit() == pit && match(cur)
- && (len = findAdvFinalize(cur, match)) > old_len);
- cur = old_cur;
- len = old_len;
- LYXERR(Debug::DEBUG, "cur_orig : " << cur_orig);
- LYXERR(Debug::DEBUG, "cur : " << cur);
- if (cur != cur_orig)
- return len;
+ if (cur)
+ cur.pos() = 0;
+ }
+ pit_changed = true;
}
- }
+ if (cur && pit_changed)
+ cur.pos() = cur.lastpos();
+ for (; cur; cur.backwardPos()) {
+ if (match(cur)) {
+ // Find the most backward consecutive match within same paragraph while searching backwards.
+ int pit = cur.pit();
+ int old_len;
+ DocIterator old_cur;
+ int len = findAdvFinalize(cur, match);
+ do {
+ old_cur = cur;
+ old_len = len;
+ cur.backwardPos();
+ LYXERR(Debug::DEBUG, "old_cur: " << old_cur << ", old_len=" << len << ", cur: " << cur);
+ } while (cur && cur.pit() == pit && match(cur)
+ && (len = findAdvFinalize(cur, match)) > old_len);
+ cur = old_cur;
+ len = old_len;
+ LYXERR(Debug::DEBUG, "cur_orig : " << cur_orig);
+ LYXERR(Debug::DEBUG, "cur : " << cur);
+ if (cur != cur_orig)
+ return len;
+ }
+ }
+ wrap_answer = frontend::Alert::prompt(
+ _("Wrap search ?"),
+ _("Beginning of document reached while searching backwards\n"
+ "\n"
+ "Continue searching from end ?"),
+ 0, 1, _("&Yes"), _("&No"));
+ cur = doc_iterator_end(&match.buf);
+ cur.backwardPos();
+ } while (wrap_answer == 0);
return 0;
}
Index: src/Paragraph.cpp
===================================================================
--- src/Paragraph.cpp (revisione 27622)
+++ src/Paragraph.cpp (copia locale)
@@ -2440,6 +2440,29 @@
}
+docstring Paragraph::stringify(pos_type beg, pos_type end, int options, OutputParams & runparams) const
+{
+ odocstringstream os;
+
+ if (beg == 0
+ && options & AS_STR_LABEL
+ && !d->params_.labelString().empty())
+ os << d->params_.labelString() << ' ';
+
+ for (pos_type i = beg; i < end; ++i) {
+ char_type const c = d->text_[i];
+ if (isPrintable(c) || c == '\t'
+ || (c == '\n' && options & AS_STR_NEWLINES))
+ os.put(c);
+ else if (c == META_INSET && options & AS_STR_INSETS) {
+ getInset(i)->plaintext(os, runparams);
+ }
+ }
+
+ return os.str();
+}
+
+
void Paragraph::setInsetOwner(Inset const * inset)
{
d->inset_owner_ = inset;
Index: src/lyxfind.h
===================================================================
--- src/lyxfind.h (revisione 27615)
+++ src/lyxfind.h (copia locale)
@@ -68,22 +68,23 @@
class FindAdvOptions {
public:
- FindAdvOptions(
- docstring const & search,
- bool casesensitive,
- bool matchword,
- bool forward,
- bool expandmacros,
- bool ignoreformat,
- bool regexp);
- FindAdvOptions() {}
- docstring search;
- bool casesensitive;
- bool matchword;
- bool forward;
- bool expandmacros;
- bool ignoreformat;
- bool regexp;
+ FindAdvOptions(
+ docstring const & search,
+ bool casesensitive,
+ bool matchword,
+ bool forward,
+ bool expandmacros,
+ bool ignoreformat,
+ bool regexp
+ );
+ FindAdvOptions() { }
+ docstring search;
+ bool casesensitive;
+ bool matchword;
+ bool forward;
+ bool expandmacros;
+ bool ignoreformat;
+ bool regexp;
};
/// Write a FindAdvOptions instance to a stringstream
@@ -105,8 +106,11 @@
** This is useful for computing opt.search from the SearchAdvDialog controller (ControlSearchAdv).
** Ideally, this should not be needed, and the opt.search field should become a Text const &.
**/
-docstring stringifyFromForSearch(FindAdvOptions const & opt,
- Buffer const & buf, DocIterator const & cur, int len = -1);
+docstring stringifyFromForSearch(
+ FindAdvOptions const & opt,
+ Buffer const & buf,
+ DocIterator const & cur,
+ int len = -1);
} // namespace lyx
Index: src/Paragraph.h
===================================================================
--- src/Paragraph.h (revisione 27615)
+++ src/Paragraph.h (copia locale)
@@ -124,6 +124,10 @@
docstring asString(pos_type beg, pos_type end,
int options = AS_STR_NONE) const;
+ /// Extract only the explicitly visible text (without any formatting),
+ /// descending into insets
+ docstring stringify(pos_type beg, pos_type end, int options, OutputParams & runparams) const;
+
///
void write(std::ostream &, BufferParams const &,
depth_type & depth) const;
