X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=util%2Fmarcdump.c;h=8b082a317d4ba28dc309e61a4415b5b7c8071933;hb=1908df105a83e97f3e404020fd9f4e2007654afd;hp=00d75960c45f12226113e75a8827e229fb19e31c;hpb=6ad80c673e9591ae178154cfe2ce002834cfdafd;p=yaz-moved-to-github.git diff --git a/util/marcdump.c b/util/marcdump.c index 00d7596..8b082a3 100644 --- a/util/marcdump.c +++ b/util/marcdump.c @@ -1,19 +1,30 @@ /* - * Copyright (c) 1995-2003, Index Data + * Copyright (C) 1995-2005, Index Data ApS * See the file LICENSE for details. - * Sebastian Hammer, Adam Dickmeiss * - * $Id: marcdump.c,v 1.23 2003-12-11 00:37:23 adam Exp $ + * $Id: marcdump.c,v 1.33 2005-08-22 20:34:23 adam Exp $ */ +#define _FILE_OFFSET_BITS 64 + #if HAVE_CONFIG_H #include #endif +#if HAVE_XML2 +#include +#include + +#include +#include + +#endif + #include #include #include #include +#include #if HAVE_LOCALE_H #include @@ -36,13 +47,64 @@ static void usage(const char *prog) { - fprintf (stderr, "Usage: %s [-c cfile] [-f from] [-t to] [-x] [-O] [-X] [-I] [-v] file...\n", + fprintf (stderr, "Usage: %s [-c cfile] [-f from] [-t to] [-x] [-O] [-X] [-e] [-I] [-v] file...\n", prog); } +#if HAVE_XML2 +void print_xpath_nodes(xmlNodeSetPtr nodes, FILE* output) { + xmlNodePtr cur; + int size; + int i; + + assert(output); + size = (nodes) ? nodes->nodeNr : 0; + + fprintf(output, "Result (%d nodes):\n", size); + for(i = 0; i < size; ++i) { + assert(nodes->nodeTab[i]); + + if(nodes->nodeTab[i]->type == XML_NAMESPACE_DECL) + { + xmlNsPtr ns; + + ns = (xmlNsPtr)nodes->nodeTab[i]; + cur = (xmlNodePtr)ns->next; + if(cur->ns) { + fprintf(output, "= namespace \"%s\"=\"%s\" for node %s:%s\n", + ns->prefix, ns->href, cur->ns->href, cur->name); + } else { + fprintf(output, "= namespace \"%s\"=\"%s\" for node %s\n", + ns->prefix, ns->href, cur->name); + } + } + else if(nodes->nodeTab[i]->type == XML_ELEMENT_NODE) + { + cur = nodes->nodeTab[i]; + if(cur->ns) { + fprintf(output, "= element node \"%s:%s\"\n", + cur->ns->href, cur->name); + } + else + { + fprintf(output, "= element node \"%s\"\n", + cur->name); + } + } + else + { + cur = nodes->nodeTab[i]; + fprintf(output, "= node \"%s\": type %d\n", cur->name, cur->type); + } + } +} +#endif + int main (int argc, char **argv) { int r; + int libxml_dom_test = 0; + int print_offset = 0; char *arg; int verbose = 0; FILE *inf; @@ -52,7 +114,7 @@ int main (int argc, char **argv) int xml = 0; FILE *cfile = 0; char *from = 0, *to = 0; - + int num = 1; #if HAVE_LOCALE_H setlocale(LC_CTYPE, ""); @@ -63,10 +125,10 @@ int main (int argc, char **argv) #endif #endif - while ((r = options("vc:xOXIf:t:", argv, argc, &arg)) != -2) + while ((r = options("pvc:xOeXIf:t:2", argv, argc, &arg)) != -2) { - int count; - no++; + int count; + no++; switch (r) { case 'f': @@ -75,34 +137,43 @@ int main (int argc, char **argv) case 't': to = arg; break; - case 'c': - if (cfile) - fclose (cfile); - cfile = fopen (arg, "w"); - break; + case 'c': + if (cfile) + fclose (cfile); + cfile = fopen(arg, "w"); + break; case 'x': xml = YAZ_MARC_SIMPLEXML; break; case 'O': xml = YAZ_MARC_OAIMARC; break; + case 'e': + xml = YAZ_MARC_XCHANGE; + break; case 'X': xml = YAZ_MARC_MARCXML; break; - case 'I': - xml = YAZ_MARC_ISO2709; - break; + case 'I': + xml = YAZ_MARC_ISO2709; + break; + case 'p': + print_offset = 1; + break; + case '2': + libxml_dom_test = 1; + break; case 0: - inf = fopen (arg, "rb"); - count = 0; - if (!inf) - { - fprintf (stderr, "%s: cannot open %s:%s\n", - prog, arg, strerror (errno)); - exit(1); - } - if (cfile) - fprintf (cfile, "char *marc_records[] = {\n"); + inf = fopen(arg, "rb"); + count = 0; + if (!inf) + { + fprintf (stderr, "%s: cannot open %s:%s\n", + prog, arg, strerror (errno)); + exit(1); + } + if (cfile) + fprintf (cfile, "char *marc_records[] = {\n"); if (1) { yaz_marc_t mt = yaz_marc_create(); @@ -117,58 +188,103 @@ int main (int argc, char **argv) "unsupported\n", from, to); exit(2); } - yaz_marc_iconv(mt, cd); + yaz_marc_iconv(mt, cd); } yaz_marc_xml(mt, xml); yaz_marc_debug(mt, verbose); while (1) { int len; - char *result; + char *result = 0; int rlen; r = fread (buf, 1, 5, inf); if (r < 5) + { + if (r && print_offset && verbose) + printf ("\n", r); break; + } + while (*buf < '0' || *buf > '9') + { + int i; + long off = ftell(inf) - 5; + if (verbose || print_offset) + printf("\n", + *buf & 0xff, *buf & 0xff, + off, off); + for (i = 0; i<4; i++) + buf[i] = buf[i+1]; + r = fread(buf+4, 1, 1, inf); + if (r < 1) + break; + } + if (r < 1) + { + if (verbose || print_offset) + printf ("\n"); + break; + } + if (print_offset) + { + long off = ftell(inf) - 5; + printf ("\n", + num, off, off); + } len = atoi_n(buf, 5); if (len < 25 || len > 100000) + { + long off = ftell(inf) - 5; + printf("Bad Length %d read at offset %ld (%lx)\n", + len, (long) off, (long) off); break; + } len = len - 5; r = fread (buf + 5, 1, len, inf); if (r < len) break; r = yaz_marc_decode_buf (mt, buf, -1, &result, &rlen); - if (r <= 0) - break; -#if 1 - fwrite (result, rlen, 1, stdout); -#else - if (!cd) + if (result) fwrite (result, rlen, 1, stdout); - else +#if HAVE_XML2 + if (r > 0 && libxml_dom_test) { - char outbuf[12]; - size_t inbytesleft = rlen; - const char *inp = result; - - while (inbytesleft) + xmlDocPtr doc = xmlParseMemory(result, rlen); + if (!doc) + fprintf(stderr, "xmLParseMemory failed\n"); + else { - size_t outbytesleft = sizeof(outbuf); - char *outp = outbuf; - size_t r = yaz_iconv (cd, (char**) &inp, - &inbytesleft, - &outp, &outbytesleft); - if (r == (size_t) (-1)) - { - int e = yaz_iconv_error(cd); - if (e != YAZ_ICONV_E2BIG) - break; + int i; + xmlXPathContextPtr xpathCtx; + xmlXPathObjectPtr xpathObj; + static const char *xpathExpr[] = { + "/record/datafield[@tag='245']/subfield[@code='a']", + "/record/datafield[@tag='100']/subfield", + "/record/datafield[@tag='245']/subfield[@code='a']", + "/record/datafield[@tag='650']/subfield", + "/record/datafield[@tag='650']", + 0}; + + xpathCtx = xmlXPathNewContext(doc); + + for (i = 0; xpathExpr[i]; i++) { + xpathObj = xmlXPathEvalExpression(BAD_CAST xpathExpr[i], xpathCtx); + if(xpathObj == NULL) { + fprintf(stderr,"Error: unable to evaluate xpath expression \"%s\"\n", xpathExpr[i]); + } + else + { + print_xpath_nodes(xpathObj->nodesetval, stdout); + xmlXPathFreeObject(xpathObj); + } } - fwrite (outbuf, outp - outbuf, 1, stdout); + xmlXPathFreeContext(xpathCtx); + xmlFreeDoc(doc); } } #endif - if (cfile) + if (r > 0 && cfile) { char *p = buf; int i; @@ -184,21 +300,24 @@ int main (int argc, char **argv) if (i < r - 1 && (i & 15) == 15) fprintf (cfile, "\"\n"); - } + } fprintf (cfile, "\"\n"); } + num++; + if (verbose) + printf("\n"); } count++; if (cd) yaz_iconv_close(cd); yaz_marc_destroy(mt); - } - if (cfile) - fprintf (cfile, "};\n"); - fclose(inf); + } + if (cfile) + fprintf (cfile, "};\n"); + fclose(inf); break; case 'v': - verbose++; + verbose++; break; default: usage(prog); @@ -206,11 +325,19 @@ int main (int argc, char **argv) } } if (cfile) - fclose (cfile); + fclose (cfile); if (!no) { usage(prog); - exit (1); + exit (1); } exit (0); } +/* + * Local variables: + * c-basic-offset: 4 + * indent-tabs-mode: nil + * End: + * vim: shiftwidth=4 tabstop=8 expandtab + */ +