X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=lily%2Fsource-file.cc;h=f895d6011eb6e62b0d0d3910c2650d818476c91c;hb=ee5919c085b841bae3f4f323a5007321f660f47c;hp=902baa0e47b4a50dec5a7a6f19062791daef3b0c;hpb=31568c504806f35aac420a394c9eab07abd9faa7;p=lilypond.git diff --git a/lily/source-file.cc b/lily/source-file.cc index 902baa0e47..f895d6011e 100644 --- a/lily/source-file.cc +++ b/lily/source-file.cc @@ -1,22 +1,32 @@ /* - source-file.cc -- implement Source_file + This file is part of LilyPond, the GNU music typesetter. - source file of the GNU LilyPond music typesetter - - (c) 1997--2006 Jan Nieuwenhuizen + Copyright (C) 1997--2010 Jan Nieuwenhuizen Han-Wen Nienhuys + + LilyPond is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + LilyPond is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with LilyPond. If not, see . */ +#if GCC_MAJOR < 4 +#define _GLIBCXX_HAVE_MBSTATE_T +#include +#endif /* GCC_MAJOR < 4 */ + #include "source-file.hh" #include "config.hh" -#if HAVE_UTF8_WCHAR_H -#include /* mbrtowc */ -#else -#include /* mbrtowc */ -#endif - #include #if HAVE_SSTREAM @@ -29,40 +39,41 @@ using namespace std; #include "file-name-map.hh" #include "international.hh" +#include "misc.hh" #include "warn.hh" void Source_file::load_stdin () { - length_ = 0; - chs_.clear (); + characters_.clear (); int c; while ((c = fgetc (stdin)) != EOF) - chs_.push_back (c); - - chs_.push_back (0); - length_ = chs_.size (); - contents_str0_ = &chs_[0]; + characters_.push_back ((char)c); } -char * -gulp_file (string filename, int *filesize) +/* + return contents of FILENAME. *Not 0-terminated!* + */ +vector +gulp_file (string filename, int desired_size) { /* "b" must ensure to open literally, avoiding text (CR/LF) conversions. */ FILE *f = fopen (filename.c_str (), "rb"); if (!f) { - warning (_f ("can't open file: `%s'", filename.c_str ())); - return 0; + warning (_f ("cannot open file: `%s'", filename.c_str ())); + + vector cxx_arr; + return cxx_arr; } fseek (f, 0, SEEK_END); int real_size = ftell (f); int read_count = real_size; - if (*filesize >= 0) - read_count = min (read_count, *filesize); + if (desired_size > 0) + read_count = min (read_count, desired_size); rewind (f); @@ -74,60 +85,75 @@ gulp_file (string filename, int *filesize) warning (_f ("expected to read %d characters, got %d", bytes_read, read_count)); fclose (f); - *filesize = bytes_read; - return str; + int filesize = bytes_read; + + vector cxx_arr; + cxx_arr.resize (filesize); + + copy (str, str + filesize, cxx_arr.begin ()); + + delete[] str; + return cxx_arr; +} + +void +Source_file::init () +{ + istream_ = 0; + line_offset_ = 0; + str_port_ = SCM_EOL; + self_scm_ = SCM_EOL; + smobify_self (); } Source_file::Source_file (string filename, string data) { + init (); + name_ = filename; - istream_ = 0; - length_ = data.length (); - contents_str0_ = string_copy (data); - pos_str0_ = c_str (); + + characters_.resize (data.length ()); + copy (data.begin (), data.end (), characters_.begin ()); + + characters_.push_back (0); + init_port (); - for (int i = 0; i < length_; i++) - if (contents_str0_[i] == '\n') - newline_locations_.push_back (contents_str0_ + i); + for (vsize i = 0; i < characters_.size (); i++) + if (characters_[i] == '\n') + newline_locations_.push_back (&characters_[0] + i); } Source_file::Source_file (string filename_string) { + init (); + name_ = filename_string; - istream_ = 0; - contents_str0_ = 0; if (filename_string == "-") load_stdin (); else { - length_ = -1; - contents_str0_ = gulp_file (filename_string, &length_); + characters_ = gulp_file (filename_string, -1); } - - pos_str0_ = c_str (); + + characters_.push_back (0); init_port (); - for (int i = 0; i < length_; i++) - if (contents_str0_[i] == '\n') - newline_locations_.push_back (contents_str0_ + i); + for (vsize i = 0; i < characters_.size (); i++) + if (characters_[i] == '\n') + newline_locations_.push_back (&characters_[0] + i); } void Source_file::init_port () { - SCM str = scm_makfrom0str (contents_str0_); + SCM str = scm_from_locale_string (c_str ()); str_port_ = scm_mkstrport (SCM_INUM0, str, SCM_OPN | SCM_RDNG, __FUNCTION__); - scm_set_port_filename_x (str_port_, scm_makfrom0str (name_.c_str ())); + scm_set_port_filename_x (str_port_, ly_string2scm (name_)); } -int -Source_file::tell () const -{ - return pos_str0_ - contents_str0_; -} istream * Source_file::get_istream () @@ -153,8 +179,8 @@ Source_file::file_line_column_string (char const *context_str0) const return " (" + _ ("position unknown") + ")"; else { - int l, ch, col; - get_counts (context_str0, &l, &ch, &col); + int l, ch, col, offset; + get_counts (context_str0, &l, &ch, &col, &offset); return name_string () + ":" + to_string (l) + ":" + to_string (col); @@ -167,13 +193,13 @@ Source_file::quote_input (char const *pos_str0) const if (!contains (pos_str0)) return " (" + _ ("position unknown") + ")"; - int l, ch, col; - get_counts (pos_str0, &l, &ch, &col); + int l, ch, col, offset; + get_counts (pos_str0, &l, &ch, &col, &offset); string line = line_string (pos_str0); - string context = line.substr (0, ch) + string context = line.substr (0, offset) + to_string ('\n') + to_string (' ', col) - + line.substr (ch, line.length()-ch); + + line.substr (offset, line.length () - offset); return context; } @@ -186,8 +212,6 @@ Source_file::name_string () const Source_file::~Source_file () { delete istream_; - istream_ = 0; - delete[] contents_str0_; } Slice @@ -235,11 +259,10 @@ void Source_file::get_counts (char const *pos_str0, int *line_number, int *line_char, - int *column) const + int *column, + int *byte_offset) const { *line_number = 0; - *line_char = 0; - *column = 0; if (!contains (pos_str0)) return; @@ -254,37 +277,13 @@ Source_file::get_counts (char const *pos_str0, string line_begin (line_start, left); char const *line_chars = line_begin.c_str (); - *column = 0; *line_char = 0; - - mbstate_t state; - - /* Initialize the state. */ - memset (&state, '\0', sizeof (state)); + *column = 0; + *byte_offset = 0; while (left > 0) { - wchar_t multibyte[2]; - - /* - FIXME, this is apparently locale dependent. - */ -#if HAVE_MBRTOWC - size_t thislen = mbrtowc (multibyte, line_chars, left, &state); -#else - size_t thislen = 1; -#endif /* !HAVE_MBRTOWC */ - - /* Stop converting at invalid character; - this can mean we have read just the first part - of a valid character. */ - if (thislen == (size_t) -1) - break; - - /* We want to handle embedded NUL bytes - but the return value is 0. Correct this. */ - if (thislen == 0) - thislen = 1; + size_t thislen = utf8_char_len (*line_chars); if (thislen == 1 && line_chars[0] == '\t') (*column) = (*column / 8 + 1) * 8; @@ -292,6 +291,14 @@ Source_file::get_counts (char const *pos_str0, (*column)++; (*line_char)++; + + /* + To have decent output in UTF-8 aware terminals, + we must keep track of the number of bytes from + the left edge of the terminal. + */ + *byte_offset += thislen; + /* Advance past this character. */ line_chars += thislen; left -= thislen; @@ -313,82 +320,69 @@ Source_file::get_line (char const *pos_str0) const if (!newline_locations_.size ()) return 1; - vsize lo = 0; - vsize hi = newline_locations_.size (); + /* this will find the '\n' character at the end of our line */ + vsize lo = lower_bound (newline_locations_, + pos_str0, + less ()); - if (newline_locations_[lo] > pos_str0) - return 1; - - if (newline_locations_[hi - 1] < pos_str0) - return hi; + /* the return value will be indexed from 1 */ + return lo + 1 + line_offset_; +} - binary_search_bounds (newline_locations_, - (char const*&)pos_str0, - default_compare, - &lo, &hi); +void +Source_file::set_line (char const *pos_str0, int line) +{ + int current_line = get_line (pos_str0); + line_offset_ += line - current_line; - if (*pos_str0 == '\n') - lo--; - return lo + 2; + assert (line == get_line (pos_str0)); } int Source_file::length () const { - return length_; + return characters_.size (); } char const * Source_file::c_str () const { - return contents_str0_; + return &characters_[0]; } -void -Source_file::set_pos (char const *pos_str0) +SCM +Source_file::get_port () const { - if (contains (pos_str0)) - pos_str0_ = pos_str0; - else - error (quote_input (pos_str0) + "invalid pos"); + return str_port_; } -char const * -Source_file::seek_str0 (int n) -{ - char const *new_str0 = c_str () + n; - if (n < 0) - new_str0 += length (); - if (contains (new_str0)) - pos_str0_ = new_str0; - else - error (quote_input (new_str0) + "seek past eof"); +/****************************************************************/ - return pos_str0_; -} +#include "ly-smobs.icc" -char const * -Source_file::forward_str0 (int n) +IMPLEMENT_SMOBS (Source_file); +IMPLEMENT_DEFAULT_EQUAL_P (Source_file); +IMPLEMENT_TYPE_P (Source_file, "ly:source-file?"); + +SCM +Source_file::mark_smob (SCM smob) { - char const *old_pos = pos_str0_; - char const *new_str0 = pos_str0_ + n; - if (contains (new_str0)) - pos_str0_ = new_str0; - else - error (quote_input (new_str0) + "forward past eof"); + Source_file *sc = (Source_file *) SCM_CELL_WORD_1 (smob); - return old_pos; + return sc->str_port_; } -string -Source_file::get_string (int n) -{ - string str = string ((char const *)forward_str0 (n), n); - return str; -} -SCM -Source_file::get_port () const +int +Source_file::print_smob (SCM smob, SCM port, scm_print_state *) { - return str_port_; + Source_file *sc = (Source_file *) SCM_CELL_WORD_1 (smob); + + scm_puts ("#name_.c_str (), port); + + /* Do not print properties, that is too much hassle. */ + scm_puts (" >", port); + return 1; } +