X-Git-Url: http://git.indexdata.com/?p=yaz-moved-to-github.git;a=blobdiff_plain;f=src%2Fmarcdisp.c;h=85acb07139e2470971e3ed442c22b58c9af8db45;hp=0ef545ef6590395254e469ff4fd2ca5c33325811;hb=8ceaeefe2e491935cba91f56007308be6e4996e6;hpb=43a9d38d20c1b1bcd1a03b2445a501d27526bd35 diff --git a/src/marcdisp.c b/src/marcdisp.c index 0ef545e..85acb07 100644 --- a/src/marcdisp.c +++ b/src/marcdisp.c @@ -1,5 +1,5 @@ /* This file is part of the YAZ toolkit. - * Copyright (C) 1995-2011 Index Data + * Copyright (C) Index Data * See the file LICENSE for details. */ @@ -20,7 +20,6 @@ #include #include #include -#include #include #include #include @@ -37,10 +36,10 @@ enum yaz_collection_state { collection_first, collection_second }; - + /** \brief node types for yaz_marc_node */ enum YAZ_MARC_NODE_TYPE -{ +{ YAZ_MARC_DATAFIELD, YAZ_MARC_CONTROLFIELD, YAZ_MARC_COMMENT, @@ -142,7 +141,7 @@ static int marc_exec_leader(const char *leader_spec, char *leader, size_t size); #if YAZ_HAVE_XML2 static int yaz_marc_write_xml_turbo_xml(yaz_marc_t mt, xmlNode **root_ptr, - const char *ns, + const char *ns, const char *format, const char *type); #endif @@ -245,16 +244,20 @@ void yaz_marc_add_datafield(yaz_marc_t mt, const char *tag, mt->subfield_pp = &n->u.datafield.subfields; } -// Magic function: adds a attribute value to the element name if it is plain characters. -// if not, and if the attribute name is not null, it will append a attribute element with the value -// if attribute name is null it will return a non-zero value meaning it couldnt handle the value. +/** \brief adds a attribute value to the element name if it is plain chars -int element_name_append_attribute_value(yaz_marc_t mt, WRBUF buffer, const char *attribute_name, char *code_data, size_t code_len) + If not, and if the attribute name is not null, it will append a + attribute element with the value if attribute name is null it will + return a non-zero value meaning it couldnt handle the value. +*/ +static int element_name_append_attribute_value( + yaz_marc_t mt, WRBUF buffer, + const char *attribute_name, char *code_data, size_t code_len) { - // TODO Map special codes to something possible for XML ELEMENT names + /* TODO Map special codes to something possible for XML ELEMENT names */ int encode = 0; - int index = 0; + size_t index = 0; int success = 0; for (index = 0; index < code_len; index++) { @@ -263,7 +266,7 @@ int element_name_append_attribute_value(yaz_marc_t mt, WRBUF buffer, const char (code_data[index] >= 'A' && code_data[index] <= 'Z'))) encode = 1; } - // Add as attribute + /* Add as attribute */ if (encode && attribute_name) wrbuf_printf(buffer, " %s=\"", attribute_name); @@ -273,7 +276,7 @@ int element_name_append_attribute_value(yaz_marc_t mt, WRBUF buffer, const char success = -1; if (encode && attribute_name) - wrbuf_printf(buffer, "\""); // return error if we couldn't handle it. + wrbuf_printf(buffer, "\""); /* return error if we couldn't handle it.*/ return success; } @@ -300,7 +303,7 @@ void yaz_marc_add_datafield_xml2(yaz_marc_t mt, char *tag_value, char *indicator n->u.datafield.indicator = indicators; n->u.datafield.subfields = 0; - // make subfield_pp the current (last one) + /* make subfield_pp the current (last one) */ mt->subfield_pp = &n->u.datafield.subfields; } @@ -339,6 +342,18 @@ void yaz_marc_add_subfield(yaz_marc_t mt, } } +static void check_ascii(yaz_marc_t mt, char *leader, int offset, + int ch_default) +{ + if (leader[offset] < ' ' || leader[offset] > 127) + { + yaz_marc_cprintf(mt, + "Leader character at offset %d is non-ASCII. " + "Setting value to '%c'", offset, ch_default); + leader[offset] = ch_default; + } +} + void yaz_marc_set_leader(yaz_marc_t mt, const char *leader_c, int *indicator_length, int *identifier_length, @@ -351,53 +366,57 @@ void yaz_marc_set_leader(yaz_marc_t mt, const char *leader_c, memcpy(leader, leader_c, 24); - if (!atoi_n_check(leader+10, 1, indicator_length)) + check_ascii(mt, leader, 5, 'a'); + check_ascii(mt, leader, 6, 'a'); + check_ascii(mt, leader, 7, 'a'); + check_ascii(mt, leader, 8, '#'); + check_ascii(mt, leader, 9, '#'); + if (!atoi_n_check(leader+10, 1, indicator_length) || *indicator_length == 0) { - yaz_marc_cprintf(mt, - "Indicator length at offset 10 should hold a digit." - " Assuming 2"); + yaz_marc_cprintf(mt, "Indicator length at offset 10 should" + " hold a number 1-9. Assuming 2"); leader[10] = '2'; *indicator_length = 2; } - if (!atoi_n_check(leader+11, 1, identifier_length)) + if (!atoi_n_check(leader+11, 1, identifier_length) || *identifier_length == 0) { - yaz_marc_cprintf(mt, - "Identifier length at offset 11 should hold a digit." - " Assuming 2"); + yaz_marc_cprintf(mt, "Identifier length at offset 11 should " + " hold a number 1-9. Assuming 2"); leader[11] = '2'; *identifier_length = 2; } if (!atoi_n_check(leader+12, 5, base_address)) { - yaz_marc_cprintf(mt, - "Base address at offsets 12..16 should hold a number." - " Assuming 0"); + yaz_marc_cprintf(mt, "Base address at offsets 12..16 should" + " hold a number. Assuming 0"); *base_address = 0; } - if (!atoi_n_check(leader+20, 1, length_data_entry)) + check_ascii(mt, leader, 17, '#'); + check_ascii(mt, leader, 18, '#'); + check_ascii(mt, leader, 19, '#'); + if (!atoi_n_check(leader+20, 1, length_data_entry) || + *length_data_entry < 3) { - yaz_marc_cprintf(mt, - "Length data entry at offset 20 should hold a digit." - " Assuming 4"); + yaz_marc_cprintf(mt, "Length data entry at offset 20 should" + " hold a number 3-9. Assuming 4"); *length_data_entry = 4; leader[20] = '4'; } - if (!atoi_n_check(leader+21, 1, length_starting)) + if (!atoi_n_check(leader+21, 1, length_starting) || *length_starting < 4) { - yaz_marc_cprintf(mt, - "Length starting at offset 21 should hold a digit." - " Assuming 5"); + yaz_marc_cprintf(mt, "Length starting at offset 21 should" + " hold a number 4-9. Assuming 5"); *length_starting = 5; leader[21] = '5'; } if (!atoi_n_check(leader+22, 1, length_implementation)) { - yaz_marc_cprintf(mt, - "Length implementation at offset 22 should hold a digit." - " Assuming 0"); + yaz_marc_cprintf(mt, "Length implementation at offset 22 should" + " hold a number. Assuming 0"); *length_implementation = 0; leader[22] = '0'; } + check_ascii(mt, leader, 23, '0'); if (mt->debug) { @@ -439,14 +458,24 @@ static size_t cdata_one_character(yaz_marc_t mt, const char *buf) size_t inbytesleft = i; size_t r = yaz_iconv(mt->iconv_cd, (char**) &inp, &inbytesleft, &outp, &outbytesleft); + yaz_iconv(mt->iconv_cd, 0, 0, &outp, &outbytesleft); if (r != (size_t) (-1)) return i; /* got a complete sequence */ } return 1; /* giving up */ } + else + { + int error = 0; + size_t no_read = 0; + (void) yaz_read_UTF8_char((const unsigned char *) buf, strlen(buf), + &no_read, &error); + if (error == 0 && no_read > 0) + return no_read; + } return 1; /* we don't know */ } - + void yaz_marc_reset(yaz_marc_t mt) { nmem_reset(mt->nmem); @@ -467,7 +496,7 @@ int yaz_marc_write_check(yaz_marc_t mt, WRBUF wr) leader = n->u.leader; break; } - + if (!leader) return -1; if (!atoi_n_check(leader+11, 1, &identifier_length)) @@ -478,7 +507,7 @@ int yaz_marc_write_check(yaz_marc_t mt, WRBUF wr) switch(n->which) { case YAZ_MARC_COMMENT: - wrbuf_iconv_write(wr, mt->iconv_cd, + wrbuf_iconv_write(wr, mt->iconv_cd, n->u.comment, strlen(n->u.comment)); wrbuf_puts(wr, "\n"); break; @@ -513,7 +542,7 @@ int yaz_marc_write_line(yaz_marc_t mt, WRBUF wr) leader = n->u.leader; break; } - + if (!leader) return -1; if (!atoi_n_check(leader+11, 1, &identifier_length)) @@ -531,12 +560,12 @@ int yaz_marc_write_line(yaz_marc_t mt, WRBUF wr) { size_t using_code_len = get_subfield_len(mt, s->code_data, identifier_length); - - wrbuf_puts (wr, mt->subfield_str); - wrbuf_iconv_write(wr, mt->iconv_cd, s->code_data, + + wrbuf_puts (wr, mt->subfield_str); + wrbuf_iconv_write(wr, mt->iconv_cd, s->code_data, using_code_len); wrbuf_iconv_puts(wr, mt->iconv_cd, " "); - wrbuf_iconv_puts(wr, mt->iconv_cd, + wrbuf_iconv_puts(wr, mt->iconv_cd, s->code_data + using_code_len); marc_iconv_reset(mt, wr); } @@ -551,7 +580,7 @@ int yaz_marc_write_line(yaz_marc_t mt, WRBUF wr) break; case YAZ_MARC_COMMENT: wrbuf_puts(wr, "("); - wrbuf_iconv_write(wr, mt->iconv_cd, + wrbuf_iconv_write(wr, mt->iconv_cd, n->u.comment, strlen(n->u.comment)); marc_iconv_reset(mt, wr); wrbuf_puts(wr, ")\n"); @@ -603,6 +632,8 @@ int yaz_marc_write_mode(yaz_marc_t mt, WRBUF wr) return yaz_marc_write_iso2709(mt, wr); case YAZ_MARC_CHECK: return yaz_marc_write_check(mt, wr); + case YAZ_MARC_JSON: + return yaz_marc_write_json(mt, wr); } return -1; } @@ -625,7 +656,7 @@ static const char *subfield_name[2] = { "subfield", "s"}; \retval -1 failure */ static int yaz_marc_write_marcxml_wrbuf(yaz_marc_t mt, WRBUF wr, - const char *ns, + const char *ns, const char *format, const char *type, int turbo) @@ -640,12 +671,12 @@ static int yaz_marc_write_marcxml_wrbuf(yaz_marc_t mt, WRBUF wr, leader = n->u.leader; break; } - + if (!leader) return -1; if (!atoi_n_check(leader+11, 1, &identifier_length)) return -1; - + if (mt->enable_collection != no_collection) { if (mt->enable_collection == collection_first) @@ -702,7 +733,7 @@ static int yaz_marc_write_marcxml_wrbuf(yaz_marc_t mt, WRBUF wr, wrbuf_iconv_write_cdata(wr, mt->iconv_cd, s->code_data, using_code_len); wrbuf_iconv_puts(wr, mt->iconv_cd, "\">"); - } + } else { element_name_append_attribute_value(mt, wr, "code", s->code_data, using_code_len); @@ -718,7 +749,7 @@ static int yaz_marc_write_marcxml_wrbuf(yaz_marc_t mt, WRBUF wr, wrbuf_puts(wr, ">\n"); } wrbuf_printf(wr, " iconv_cd, n->u.datafield.tag, strlen(n->u.datafield.tag)); @@ -735,7 +766,7 @@ static int yaz_marc_write_marcxml_wrbuf(yaz_marc_t mt, WRBUF wr, } else { - //TODO convert special + /* TODO convert special */ wrbuf_iconv_write_cdata(wr, mt->iconv_cd, n->u.controlfield.tag, strlen(n->u.controlfield.tag)); wrbuf_iconv_puts(wr, mt->iconv_cd, ">"); @@ -745,7 +776,7 @@ static int yaz_marc_write_marcxml_wrbuf(yaz_marc_t mt, WRBUF wr, strlen(n->u.controlfield.data)); marc_iconv_reset(mt, wr); wrbuf_printf(wr, "iconv_cd, n->u.controlfield.tag, strlen(n->u.controlfield.tag)); @@ -769,7 +800,7 @@ static int yaz_marc_write_marcxml_wrbuf(yaz_marc_t mt, WRBUF wr, } static int yaz_marc_write_marcxml_ns(yaz_marc_t mt, WRBUF wr, - const char *ns, + const char *ns, const char *format, const char *type, int turbo) @@ -848,7 +879,7 @@ void add_marc_datafield_turbo_xml(yaz_marc_t mt, struct yaz_marc_node *n, struct yaz_marc_subfield *s; WRBUF subfield_name = wrbuf_alloc(); - //TODO consider if safe + /* TODO consider if safe */ char field[10]; field[0] = 'd'; strncpy(field + 1, n->u.datafield.tag, 3); @@ -862,7 +893,7 @@ void add_marc_datafield_turbo_xml(yaz_marc_t mt, struct yaz_marc_node *n, { char ind_str[6]; char ind_val[2]; - + ind_val[0] = n->u.datafield.indicator[i]; ind_val[1] = '\0'; sprintf(ind_str, "%s%d", indicator_name[1], i+1); @@ -878,7 +909,7 @@ void add_marc_datafield_turbo_xml(yaz_marc_t mt, struct yaz_marc_node *n, wrbuf_rewind(wr_cdata); wrbuf_iconv_puts(wr_cdata, mt->iconv_cd, s->code_data + using_code_len); marc_iconv_reset(mt, wr_cdata); - + wrbuf_rewind(subfield_name); wrbuf_puts(subfield_name, "s"); not_written = element_name_append_attribute_value(mt, subfield_name, 0, s->code_data, using_code_len) != 0; @@ -887,7 +918,7 @@ void add_marc_datafield_turbo_xml(yaz_marc_t mt, struct yaz_marc_node *n, BAD_CAST wrbuf_cstr(wr_cdata)); if (not_written) { - // Generate code attribute value and add + /* Generate code attribute value and add */ wrbuf_rewind(wr_cdata); wrbuf_iconv_write(wr_cdata, mt->iconv_cd,s->code_data, using_code_len); xmlNewProp(ptr_subfield, BAD_CAST "code", BAD_CAST wrbuf_cstr(wr_cdata)); @@ -897,7 +928,7 @@ void add_marc_datafield_turbo_xml(yaz_marc_t mt, struct yaz_marc_node *n, } static int yaz_marc_write_xml_turbo_xml(yaz_marc_t mt, xmlNode **root_ptr, - const char *ns, + const char *ns, const char *format, const char *type) { @@ -914,7 +945,7 @@ static int yaz_marc_write_xml_turbo_xml(yaz_marc_t mt, xmlNode **root_ptr, leader = n->u.leader; break; } - + if (!leader) return -1; if (!atoi_n_check(leader+11, 1, &identifier_length)) @@ -939,7 +970,7 @@ static int yaz_marc_write_xml_turbo_xml(yaz_marc_t mt, xmlNode **root_ptr, char field[10]; field[0] = 'c'; field[4] = '\0'; - + switch(n->which) { case YAZ_MARC_DATAFIELD: @@ -949,7 +980,7 @@ static int yaz_marc_write_xml_turbo_xml(yaz_marc_t mt, xmlNode **root_ptr, wrbuf_rewind(wr_cdata); wrbuf_iconv_puts(wr_cdata, mt->iconv_cd, n->u.controlfield.data); marc_iconv_reset(mt, wr_cdata); - + strncpy(field + 1, n->u.controlfield.tag, 3); ptr = xmlNewTextChild(record_ptr, ns_record, BAD_CAST field, @@ -960,13 +991,9 @@ static int yaz_marc_write_xml_turbo_xml(yaz_marc_t mt, xmlNode **root_ptr, xmlAddChild(record_ptr, ptr); break; case YAZ_MARC_LEADER: - { - char *field = "leader"; - field = "l"; - xmlNewTextChild(record_ptr, ns_record, BAD_CAST field, + xmlNewTextChild(record_ptr, ns_record, BAD_CAST "l", BAD_CAST n->u.leader); - } - break; + break; } } wrbuf_destroy(wr_cdata); @@ -975,7 +1002,7 @@ static int yaz_marc_write_xml_turbo_xml(yaz_marc_t mt, xmlNode **root_ptr, int yaz_marc_write_xml(yaz_marc_t mt, xmlNode **root_ptr, - const char *ns, + const char *ns, const char *format, const char *type) { @@ -992,7 +1019,7 @@ int yaz_marc_write_xml(yaz_marc_t mt, xmlNode **root_ptr, leader = n->u.leader; break; } - + if (!leader) return -1; if (!atoi_n_check(leader+11, 1, &identifier_length)) @@ -1058,11 +1085,11 @@ int yaz_marc_write_xml(yaz_marc_t mt, xmlNode **root_ptr, wrbuf_rewind(wr_cdata); wrbuf_iconv_puts(wr_cdata, mt->iconv_cd, n->u.controlfield.data); marc_iconv_reset(mt, wr_cdata); - + ptr = xmlNewTextChild(record_ptr, ns_record, BAD_CAST "controlfield", BAD_CAST wrbuf_cstr(wr_cdata)); - + xmlNewProp(ptr, BAD_CAST "tag", BAD_CAST n->u.controlfield.tag); break; case YAZ_MARC_COMMENT: @@ -1093,11 +1120,11 @@ int yaz_marc_write_iso2709(yaz_marc_t mt, WRBUF wr) const char *leader = 0; WRBUF wr_dir, wr_head, wr_data_tmp; int base_address; - + for (n = mt->nodes; n; n = n->next) if (n->which == YAZ_MARC_LEADER) leader = n->u.leader; - + if (!leader) return -1; if (!atoi_n_check(leader+10, 1, &indicator_length)) @@ -1140,7 +1167,7 @@ int yaz_marc_write_iso2709(yaz_marc_t mt, WRBUF wr) wrbuf_printf(wr_dir, "%.3s", n->u.controlfield.tag); wrbuf_rewind(wr_data_tmp); - wrbuf_iconv_puts(wr_data_tmp, mt->iconv_cd, + wrbuf_iconv_puts(wr_data_tmp, mt->iconv_cd, n->u.controlfield.data); marc_iconv_reset(mt, wr_data_tmp); wrbuf_iconv_putchar(wr_data_tmp, mt->iconv_cd, ' ');/* field sep */ @@ -1175,7 +1202,7 @@ int yaz_marc_write_iso2709(yaz_marc_t mt, WRBUF wr) wrbuf_printf(wr_head, "%05d", base_address); /* from "original" leader */ wrbuf_write(wr_head, leader+17, 7); - + wrbuf_write(wr, wrbuf_buf(wr_head), 24); wrbuf_write(wr, wrbuf_buf(wr_dir), wrbuf_len(wr_dir)); wrbuf_destroy(wr_head); @@ -1189,8 +1216,7 @@ int yaz_marc_write_iso2709(yaz_marc_t mt, WRBUF wr) switch(n->which) { case YAZ_MARC_DATAFIELD: - wrbuf_printf(wr, "%.*s", indicator_length, - n->u.datafield.indicator); + wrbuf_write(wr, n->u.datafield.indicator, indicator_length); for (s = n->u.datafield.subfields; s; s = s->next) { wrbuf_putc(wr, ISO2709_IDFS); @@ -1214,6 +1240,91 @@ int yaz_marc_write_iso2709(yaz_marc_t mt, WRBUF wr) return 0; } +int yaz_marc_write_json(yaz_marc_t mt, WRBUF w) +{ + int identifier_length; + struct yaz_marc_node *n; + const char *leader = 0; + int first = 1; + + wrbuf_puts(w, "{\n"); + for (n = mt->nodes; n; n = n->next) + if (n->which == YAZ_MARC_LEADER) + leader = n->u.leader; + + if (!leader) + return -1; + + if (!atoi_n_check(leader+11, 1, &identifier_length)) + return -1; + + wrbuf_puts(w, "\t\"leader\":\""); + wrbuf_json_puts(w, leader); + wrbuf_puts(w, "\",\n"); + wrbuf_puts(w, "\t\"fields\":\n\t[\n"); + + for (n = mt->nodes; n; n = n->next) + { + struct yaz_marc_subfield *s; + const char *sep = ""; + switch (n->which) + { + case YAZ_MARC_LEADER: + case YAZ_MARC_COMMENT: + break; + case YAZ_MARC_CONTROLFIELD: + if (first) + first = 0; + else + wrbuf_puts(w, ",\n"); + wrbuf_puts(w, "\t\t{\n\t\t\t\""); + wrbuf_iconv_json_puts(w, mt->iconv_cd, n->u.controlfield.tag); + wrbuf_puts(w, "\":\""); + wrbuf_iconv_json_puts(w, mt->iconv_cd, n->u.controlfield.data); + wrbuf_puts(w, "\"\n\t\t}"); + break; + case YAZ_MARC_DATAFIELD: + if (first) + first = 0; + else + wrbuf_puts(w, ",\n"); + + wrbuf_puts(w, "\t\t{\n\t\t\t\""); + wrbuf_json_puts(w, n->u.datafield.tag); + wrbuf_puts(w, "\":\n\t\t\t{\n\t\t\t\t\"subfields\":\n\t\t\t\t[\n"); + for (s = n->u.datafield.subfields; s; s = s->next) + { + size_t using_code_len = get_subfield_len(mt, s->code_data, + identifier_length); + wrbuf_puts(w, sep); + sep = ",\n"; + wrbuf_puts(w, "\t\t\t\t\t{\n\t\t\t\t\t\t\""); + wrbuf_iconv_json_write(w, mt->iconv_cd, + s->code_data, using_code_len); + wrbuf_puts(w, "\":\""); + wrbuf_iconv_json_puts(w, mt->iconv_cd, + s->code_data + using_code_len); + wrbuf_puts(w, "\"\n\t\t\t\t\t}"); + } + wrbuf_puts(w, "\n\t\t\t\t]"); + if (n->u.datafield.indicator[0]) + { + int i; + for (i = 0; n->u.datafield.indicator[i]; i++) + { + wrbuf_printf(w, ",\n\t\t\t\t\"ind%d\":\"%c\"", i + 1, + n->u.datafield.indicator[i]); + } + } + wrbuf_puts(w, "\n\t\t\t}\n"); + wrbuf_puts(w, "\n\t\t}"); + break; + } + } + wrbuf_puts(w, "\n\t]\n"); + wrbuf_puts(w, "}\n"); + return 0; +} int yaz_marc_decode_wrbuf(yaz_marc_t mt, const char *buf, int bsize, WRBUF wr) { @@ -1307,7 +1418,7 @@ static int marc_exec_leader(const char *leader_spec, char *leader, size_t size) { const char *vp = strchr(val+1, '\''); size_t len; - + if (!vp) return -1; len = vp-val-1; @@ -1333,7 +1444,7 @@ static int marc_exec_leader(const char *leader_spec, char *leader, size_t size) int yaz_marc_decode_formatstr(const char *arg) { - int mode = -1; + int mode = -1; if (!strcmp(arg, "marc")) mode = YAZ_MARC_ISO2709; if (!strcmp(arg, "marcxml")) @@ -1344,6 +1455,8 @@ int yaz_marc_decode_formatstr(const char *arg) mode = YAZ_MARC_XCHANGE; if (!strcmp(arg, "line")) mode = YAZ_MARC_LINE; + if (!strcmp(arg, "json")) + mode = YAZ_MARC_JSON; return mode; } @@ -1352,6 +1465,16 @@ void yaz_marc_write_using_libxml2(yaz_marc_t mt, int enable) mt->write_using_libxml2 = enable; } +int yaz_marc_check_marc21_coding(const char *charset, + const char *marc_buf, int sz) +{ + if ((!yaz_matchstr(charset, "MARC8?") || + !yaz_matchstr(charset, "MARC8")) && marc_buf && sz > 25 + && marc_buf[9] == 'a') + return 1; + return 0; +} + /* * Local variables: * c-basic-offset: 4