X-Git-Url: https://git.distorted.org.uk/~mdw/sgt/halibut/blobdiff_plain/f4551933f1b39e3d017ce964cd3284193fb34cbe..8f664e7e91c918cd13248f6b684580c4dd2cdb31:/main.c diff --git a/main.c b/main.c index 4dec663..dae6581 100644 --- a/main.c +++ b/main.c @@ -3,6 +3,7 @@ */ #include +#include #include #include #include "halibut.h" @@ -11,20 +12,29 @@ static void dbg_prtsource(paragraph *sourceform); static void dbg_prtwordlist(int level, word *w); static void dbg_prtkws(keywordlist *kws); +static const struct pre_backend { + void *(*func)(paragraph *, keywordlist *, indexdata *); + int bitfield; +} pre_backends[] = { + {paper_pre_backend, 0x0001} +}; + static const struct backend { char *name; - void (*func)(paragraph *, keywordlist *, indexdata *); + void (*func)(paragraph *, keywordlist *, indexdata *, void *); paragraph *(*filename)(char *filename); - int bitfield; + int bitfield, prebackend_bitfield; } backends[] = { - {"text", text_backend, text_config_filename, 0x0001}, - {"xhtml", xhtml_backend, xhtml_config_filename, 0x0002}, - {"html", xhtml_backend, xhtml_config_filename, 0x0002}, - {"hlp", whlp_backend, whlp_config_filename, 0x0004}, - {"whlp", whlp_backend, whlp_config_filename, 0x0004}, - {"winhelp", whlp_backend, whlp_config_filename, 0x0004}, - {"man", man_backend, man_config_filename, 0x0008}, - {"info", info_backend, info_config_filename, 0x0010}, + {"text", text_backend, text_config_filename, 0x0001, 0}, + {"xhtml", html_backend, html_config_filename, 0x0002, 0}, + {"html", html_backend, html_config_filename, 0x0002, 0}, + {"hlp", whlp_backend, whlp_config_filename, 0x0004, 0}, + {"whlp", whlp_backend, whlp_config_filename, 0x0004, 0}, + {"winhelp", whlp_backend, whlp_config_filename, 0x0004, 0}, + {"man", man_backend, man_config_filename, 0x0008, 0}, + {"info", info_backend, info_config_filename, 0x0010, 0}, + {"ps", ps_backend, ps_config_filename, 0x0020, 0x0001}, + {"pdf", pdf_backend, pdf_config_filename, 0x0040, 0x0001}, }; int main(int argc, char **argv) { @@ -33,18 +43,34 @@ int main(int argc, char **argv) { int nogo; int errs; int reportcols; + int list_fonts; + int input_charset; int debug; - int backendbits; + int backendbits, prebackbits; int k, b; paragraph *cfg, *cfg_tail; + void *pre_backend_data[16]; + + /* + * Use the specified locale everywhere. It'll be used for + * output of error messages, and as the default character set + * for input files if one is not explicitly specified. + * + * However, we need to use standard numeric formatting for + * output of things like PDF. + */ + setlocale(LC_ALL, ""); + setlocale(LC_NUMERIC, "C"); /* * Set up initial (default) parameters. */ - infiles = mknewa(char *, argc); + infiles = snewn(argc, char *); nfiles = 0; nogo = errs = FALSE; reportcols = 0; + list_fonts = 0; + input_charset = CS_ASCII; debug = 0; backendbits = 0; cfg = cfg_tail = NULL; @@ -59,7 +85,7 @@ int main(int argc, char **argv) { */ while (--argc) { char *p = *++argv; - if (*p == '-') { + if (*p == '-' && p[1]) { /* * An option. */ @@ -100,6 +126,17 @@ int main(int argc, char **argv) { } if (k < (int)lenof(backends)) { /* do nothing */; + } else if (!strcmp(opt, "-input-charset")) { + if (!val) { + errs = TRUE, error(err_optnoarg, opt); + } else { + int charset = charset_from_localenc(val); + if (charset == CS_NONE) { + errs = TRUE, error(err_cmdcharset, val); + } else { + input_charset = charset; + } + } } else if (!strcmp(opt, "-help")) { help(); nogo = TRUE; @@ -110,6 +147,11 @@ int main(int argc, char **argv) { !strcmp(opt, "-license")) { licence(); nogo = TRUE; + } else if (!strcmp(opt, "-list-charsets")) { + listcharsets(); + nogo = TRUE; + } else if (!strcmp(opt, "-list-fonts")) { + list_fonts = TRUE; } else if (!strcmp(opt, "-precise")) { reportcols = 1; } else { @@ -171,36 +213,30 @@ int main(int argc, char **argv) { * into a config paragraph. */ { - wchar_t *keywords; - char *q; - wchar_t *u; + char *s = dupstr(p), *q, *r; paragraph *para; - keywords = mknewa(wchar_t, 2+strlen(p)); - - u = keywords; - q = p; + para = cmdline_cfg_new(); + q = r = s; while (*q) { if (*q == ':') { - *u++ = L'\0'; + *r = '\0'; + /* XXX ad-hoc diagnostic */ + if (!strcmp(s, "input-charset")) + error(err_futileopt, "Cinput-charset", + "; use --input-charset"); + cmdline_cfg_add(para, s); + r = s; } else { if (*q == '\\' && q[1]) q++; - /* FIXME: lacks charset flexibility */ - *u++ = *q; + *r++ = *q; } q++; } - *u = L'\0'; - - para = mknew(paragraph); - memset(para, 0, sizeof(*para)); - para->type = para_Config; - para->keyword = keywords; - para->next = NULL; - para->fpos.filename = ""; - para->fpos.line = para->fpos.col = -1; + *r = '\0'; + cmdline_cfg_add(para, s); if (cfg_tail) cfg_tail->next = para; @@ -228,7 +264,10 @@ int main(int argc, char **argv) { /* * A non-option argument. */ - infiles[nfiles++] = p; + if (!strcmp(p, "-")) + infiles[nfiles++] = NULL; /* special case: read stdin */ + else + infiles[nfiles++] = p; } } @@ -240,7 +279,7 @@ int main(int argc, char **argv) { /* * Do the work. */ - if (nfiles == 0) { + if (nfiles == 0 && !list_fonts) { error(err_noinput); usage(); exit(EXIT_FAILURE); @@ -260,10 +299,15 @@ int main(int argc, char **argv) { in.pushback = NULL; in.reportcols = reportcols; in.stack = NULL; + in.defcharset = input_charset; idx = make_index(); sourceform = read_input(&in, idx); + if (list_fonts) { + listfonts(); + exit(EXIT_SUCCESS); + } if (!sourceform) exit(EXIT_FAILURE); @@ -284,8 +328,6 @@ int main(int argc, char **argv) { sfree(in.pushback); - mark_attr_ends(sourceform); - sfree(infiles); keywords = get_keywords(sourceform); @@ -300,6 +342,20 @@ int main(int argc, char **argv) { build_index(idx); + /* + * Set up attr_First / attr_Last / attr_Always, in the main + * document and in the index entries. + */ + for (p = sourceform; p; p = p->next) + mark_attr_ends(p->words); + { + int i; + indexentry *entry; + + for (i = 0; (entry = index234(idx->entries, i)) != NULL; i++) + mark_attr_ends(entry->text); + } + if (debug) { index_debug(idx); dbg_prtkws(keywords); @@ -307,13 +363,39 @@ int main(int argc, char **argv) { } /* + * Select and run the pre-backends. + */ + prebackbits = 0; + for (k = 0; k < (int)lenof(backends); k++) + if (backendbits == 0 || (backendbits & backends[k].bitfield)) + prebackbits |= backends[k].prebackend_bitfield; + for (k = 0; k < (int)lenof(pre_backends); k++) + if (prebackbits & pre_backends[k].bitfield) { + assert(k < (int)lenof(pre_backend_data)); + pre_backend_data[k] = + pre_backends[k].func(sourceform, keywords, idx); + } + + /* * Run the selected set of backends. */ for (k = b = 0; k < (int)lenof(backends); k++) if (b != backends[k].bitfield) { b = backends[k].bitfield; - if (backendbits == 0 || (backendbits & b)) - backends[k].func(sourceform, keywords, idx); + if (backendbits == 0 || (backendbits & b)) { + void *pbd = NULL; + int pbb = backends[k].prebackend_bitfield; + int m; + + for (m = 0; m < (int)lenof(pre_backends); m++) + if (pbb & pre_backends[m].bitfield) { + assert(m < (int)lenof(pre_backend_data)); + pbd = pre_backend_data[m]; + break; + } + + backends[k].func(sourceform, keywords, idx, pbd); + } } free_para_list(sourceform); @@ -388,6 +470,8 @@ static void dbg_prtwordlist(int level, word *w) { printf("\""); } else printf("(no text)"); + if (w->breaks) + printf(" [breaks]"); if (w->alt) { printf(" alt = {\n"); dbg_prtwordlist(level+1, w->alt);