Merge branch 'master' of github.com:apertium/lttoolbox

lttoolbox/compression.cc

@@ -307,8 +307,8 @@ Compression::long_multibyte_write(const double& value, FILE *output)
 {
  int exp = 0;
 
- unsigned int mantissa = static_cast<unsigned int>(0x40000000 * frexp(value, &exp));
- unsigned int exponent = static_cast<unsigned int>(exp);
+ unsigned int mantissa = static_cast<unsigned int>(static_cast<int>(0x40000000 * frexp(value, &exp)));
+ unsigned int exponent = static_cast<unsigned int>(static_cast<int>(exp));
 
  if(mantissa < 0x04000000)
  {
@@ -344,8 +344,8 @@ Compression::long_multibyte_write(const double& value, ostream &output)
 {
  int exp = 0;
 
- unsigned int mantissa = static_cast<unsigned int>(0x40000000 * frexp(value, &exp));
- unsigned int exponent = static_cast<unsigned int>(exp);
+ unsigned int mantissa = static_cast<unsigned int>(static_cast<int>(0x40000000 * frexp(value, &exp)));
+ unsigned int exponent = static_cast<unsigned int>(static_cast<int>(exp));
 
  if(mantissa < 0x04000000)
  {

lttoolbox/fst_processor.cc

@@ -882,22 +882,19 @@ FSTProcessor::lsx(FILE *input, FILE *output)
 
  alive_states.push_back(initial_state);
 
- while(!feof(input))
- {
- int val = fgetwc_unlocked(input);
+ int val = -1;
 
- if (val == 0) {
- blankqueue.push(blank);
- break;
- }
+ while(!feof(input) && val != 0)
+ {
+ val = fgetwc_unlocked(input);
 
  if(val == L'+' && isEscaped(val) && !outOfWord)
  {
  val = L'$';
  plus_thing = true;
  }
 
- if((val == L'^' && isEscaped(val) && outOfWord) || feof(input))
+ if((val == L'^' && isEscaped(val) && outOfWord) || feof(input) || val == 0)
  {
  blankqueue.push(blank);
 
@@ -940,15 +937,15 @@ FSTProcessor::lsx(FILE *input, FILE *output)
  continue;
  }
 
- //wcerr << L"\n[!] " << (wchar_t)val << L" ||| " << outOfWord << endl;
+ // wcerr << L"\n[!] " << (wchar_t)val << L" ||| " << outOfWord << endl;
 
  if(outOfWord)
  {
  blank += val;
  continue;
  }
 
- if((feof(input) || val == L'$') && !outOfWord) // && isEscaped(val)
+ if((val == 0 || feof(input) || val == L'$') && !outOfWord) // && isEscaped(val)
  {
  new_states.clear();
  for(vector<State>::const_iterator it = alive_states.begin(); it != alive_states.end(); it++)
@@ -2150,11 +2147,14 @@ FSTProcessor::intergeneration(FILE *input, FILE *output)
  }
  else
  {
- if (isEscaped(val))
+ if(val != L'\0')
  {
- fputwc_unlocked(L'\\', output);
+ if (isEscaped(val))
+ {
+ fputwc_unlocked(L'\\', output);
+ }
+ fputwc_unlocked(val, output);
  }
- fputwc_unlocked(val, output);
  }
  }
  else
@@ -2166,56 +2166,9 @@ FSTProcessor::intergeneration(FILE *input, FILE *output)
  bool uppercase = source.size() > 1 && firstupper && iswupper(source[2]);
  target = current_state.filterFinals(all_finals, alphabet,
  empty_escaped_chars,
+ displayWeightsMode, maxAnalyses, maxWeightClasses,
  uppercase, firstupper, 0);
 
- // case of the beggining of the next word
-
- wstring mybuf = L"";
- for (size_t i = source.size(); i > 0; --i)
- {
- if (!isalpha(source[i - 1]))
- {
- break;
- }
- else
- {
- mybuf = source[i - 1] + mybuf;
- }
- }
-
- if (mybuf.size() > 0)
- {
- bool myfirstupper = iswupper(mybuf[0]);
- bool myuppercase = mybuf.size() > 1 && iswupper(mybuf[1]);
-
- for (size_t i = target.size(); i > 0; --i)
- {
- if (!isalpha(target[i - 1]))
- {
- if (myfirstupper && i != target.size())
- {
- target[i] = towupper(target[i]);
- }
- else
- {
- target[i] = towlower(target[i]);
- }
- break;
- }
- else
- {
- if (myuppercase)
- {
- target[i - 1] = towupper(target[i - 1]);
- }
- else
- {
- target[i - 1] = towlower(target[i - 1]);
- }
- }
- }
- }
-
  last = input_buffer.getPos();
  }
 

lttoolbox/lt-comp.1

@@ -1,97 +1,87 @@
-.TH lt-comp 1 2006-03-08 "" ""
-.SH NAME
-lt-comp \- This application is part of the lexical processing modules
-and tools (
-.B lttoolbox
-)
-.PP
-This tool is part of the apertium machine translation
-architecture: \fBhttp://www.apertium.org\fR.
-.SH SYNOPSIS
-.B lt-comp
-[
-.B \-a \fR|
-.B \-v \fR|
-.B \-l \fR|
-.B \-r \fR|
-.B \-h
-]
-[
-.B lr \fR|
-.B rl
-] dictionary_file output_file [acx_file]
-.PP
-.B lt-comp
-[
-.B \-\-alt \fR|
-.B \-\-var \fR|
-.B \-\-var\-left \fR|
-.B \-\-var\-right \fR|
-.B \-\-help
-]
-[
-.B lr \fR|
-.B rl
-] dictionary_file output_file [acx_file]
-.PP
-.SH DESCRIPTION
-.BR lt-comp
-Is the application responsible of compiling dictionaries used by
-\fBlt-proc\fR in \fIApertium\fR into a compact and efficient
-representation (a class of finite-state transducers called augmented
-letter transducers).
-.PP
-.SH OPTIONS
-.TP
-.B \-a, \-\-alt
-Sets the value of the \fIalt\fR attribute to use in compilation.
-
+.Dd March 8, 2006
+.Dt LT-COMP 1
+.Os Apertium
+.Sh NAME
+.Nm lt-comp
+.Nd augmented letter transducer compiler for Apertium
+.Sh SYNOPSIS
+.Nm lt-comp
+.Op Fl a | v | l | r | h
+.Cm lr | rl
+.Ar dictionary_file
+.Ar output_file
+.Op Ar acx_file
+.Sh DESCRIPTION
+.Nm lt-comp
+is the application responsible for compiling dictionaries used by
+.Xr lt-proc 1
+in Apertium into a compact and efficient representation
+(a class of finite-state transducers called augmented letter transducers).
+.Sh OPTIONS
+.Bl -tag -width Ds
+.It Fl a , Fl Fl alt
+Sets the value of the
+.Sy alt
+attribute to use in compilation.
+.Pp
 Note that if no value is set, all entries containing an \fIalt\fR
 attribute are omitted.
-.TP
-.B \-v, \-\-var
-Sets the value of the \fIv\fR attribute to use in compilation.
-This should only be used with monodixes; for bidixes, see \-l and \-r.
-
-Note that if no value is set, all entries containing a \fIv\fR
-attribute are considered to be \fIleft-to-right\fR.
-.TP
-.B \-l, \-\-var\-left
-Sets the value of the \fIvl\fR attribute for use in compilation of bidixes.
-"Left" here refers to the side of the dictionary, so this option is only valid
-in \fIrl\fR mode.
-.TP
-.B \-r, \-\-var\-right
-Sets the value of the \fIvr\fR attribute for use in compilation of bidixes.
-"Right" here refers to the side of the dictionary, so this option is only valid
-in \fIlr\fR mode.
-.TP
-.B \-h, \-\-help
-Prints a short help message
-.TP
-.B lr
+.It Fl v , Fl Fl var
+Sets the value of the
+.Sy v
+attribute to use in compilation.
+This should only be used with monodixes; for bidixes, see
+.Fl l
+and
+.Fl r .
+.Pp
+Note that if no value is set, all entries containing a
+.Sy v
+attribute are considered to be
+.Em left-to-right .
+.It Fl l , Fl Fl var-left
+Sets the value of the
+.Sy vl
+attribute for use in compilation of bidixes.
+.Dq Left
+here refers to the side of the dictionary, so this option is only valid in
+.Cm rl
+mode.
+.It Fl r , Fl Fl var-right
+Sets the value of the
+.Sy vr
+attribute for use in compilation of bidixes.
+.Dq Right
+here refers to the side of the dictionary, so this option is only valid in
+.Cm lr
+mode.
+.It Fl h , Fl Fl help
+Prints a short help message.
+.It Cm lr
 The resulting transducer will process dictionary entries
-\fIleft-to-right\fR.
-.TP
-.B rl
+.Em left-to-right .
+.It Cm rl
 The resulting transducer will process dictionary entries
-\fIright-to-left\fR.
-.SH FILES
-.B dictionary_file
+.Em right-to-left .
+.El
+.Sh FILES
+.Bl -tag -width Ds
+.It Ar dictionary_file
 The input dictionary.
-.PP
-.B output_file
+.It Ar output_file
 The compiled dictionary (a finite state transducer).
-.PP
-.B acx_file
+.It Ar acx_file
 Optional XML file of equivalent characters in monodices.
-
-.SH SEE ALSO
-.I lt-proc\fR(1),
-.I lt-expand\fR(1),
-.I apertium-tagger\fR(1),
-.I apertium\fR(1).
-.SH BUGS
-Lots of...lurking in the dark and waiting for you!
-.SH AUTHOR
-(c) 2005,2006 Universitat d'Alacant / Universidad de Alicante.
+.El
+.Sh SEE ALSO
+.Xr apertium 1 ,
+.Xr apertium-tagger 1 ,
+.Xr lt-expand 1 ,
+.Xr lt-proc 1
+.Sh COPYRIGHT
+Copyright \(co 2005, 2006 Universitat d'Alacant / Universidad de Alicante.
+This is free software.
+You may redistribute copies of it under the terms of
+.Lk https://www.gnu.org/licenses/gpl.html the GNU General Public License .
+.Sh BUGS
+Many... lurking in the dark and waiting for you!