X-Git-Url: http://lists.indexdata.dk/cgi-bin?a=blobdiff_plain;f=recctrl%2Fxslt.c;h=6331754f307a5f9491b3c15db1a3b6d7559148ef;hb=cae6de97d03369a9429024f9d4abc9f1e5fbf3bb;hp=9fe383a161c7fbb576c21bb0929973a28e71334b;hpb=b912fcbe609de53861c5cf9b5196986dd97a1462;p=idzebra-moved-to-github.git diff --git a/recctrl/xslt.c b/recctrl/xslt.c index 9fe383a..6331754 100644 --- a/recctrl/xslt.c +++ b/recctrl/xslt.c @@ -1,4 +1,4 @@ -/* $Id: xslt.c,v 1.7 2005-06-01 07:32:46 adam Exp $ +/* $Id: xslt.c,v 1.12 2005-06-24 13:45:54 adam Exp $ Copyright (C) 1995-2005 Index Data ApS @@ -28,6 +28,7 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #include #include +#include #include #include @@ -40,6 +41,7 @@ struct filter_schema { const char *stylesheet; struct filter_schema *next; const char *default_schema; + const char *include_snippet; xsltStylesheetPtr stylesheet_xsp; }; @@ -53,9 +55,19 @@ struct filter_info { xmlTextReaderPtr reader; }; -#define ZEBRA_INDEX_NS "http://indexdata.dk/zebra/indexing/1" -#define ZEBRA_SCHEMA_IDENTITY_NS "http://indexdata.dk/zebra/identity/1" -static const char *zebra_index_ns = ZEBRA_INDEX_NS; +#define ZEBRA_SCHEMA_XSLT_NS "http://indexdata.dk/zebra/xslt/1" + +static const char *zebra_xslt_ns = ZEBRA_SCHEMA_XSLT_NS; + +static void set_param_xml(const char **params, const char *name, + const char *value, ODR odr) +{ + while (*params) + params++; + params[0] = name; + params[1] = value; + params[2] = 0; +} static void set_param_str(const char **params, const char *name, const char *value, ODR odr) @@ -82,6 +94,32 @@ static void set_param_int(const char **params, const char *name, } +int zebra_xmlInputMatchCallback (char const *filename) +{ + yaz_log(YLOG_LOG, "match %s", filename); + return 0; +} + + +void * zebra_xmlInputOpenCallback (char const *filename) +{ + return 0; +} + +int zebra_xmlInputReadCallback (void * context, char * buffer, int len) +{ + return 0; +} + +int zebra_xmlInputCloseCallback (void * context) +{ + return 0; +} + + + + + static void *filter_init_xslt(Res res, RecType recType) { struct filter_info *tinfo = (struct filter_info *) xmalloc(sizeof(*tinfo)); @@ -92,6 +130,14 @@ static void *filter_init_xslt(Res res, RecType recType) tinfo->odr = odr_createmem(ODR_ENCODE); tinfo->doc = 0; tinfo->schemas = 0; + +#if 0 + xmlRegisterDefaultInputCallbacks(); + xmlRegisterInputCallbacks(zebra_xmlInputMatchCallback, + zebra_xmlInputOpenCallback, + zebra_xmlInputReadCallback, + zebra_xmlInputCloseCallback); +#endif return tinfo; } @@ -158,6 +204,7 @@ static ZEBRA_RES create_schemas(struct filter_info *tinfo, const char *fname) schema->default_schema = 0; schema->next = tinfo->schemas; schema->stylesheet_xsp = 0; + schema->include_snippet = 0; tinfo->schemas = schema; for (attr = ptr->properties; attr; attr = attr->next) { @@ -165,6 +212,7 @@ static ZEBRA_RES create_schemas(struct filter_info *tinfo, const char *fname) attr_content(attr, "name", &schema->name); attr_content(attr, "stylesheet", &schema->stylesheet); attr_content(attr, "default", &schema->default_schema); + attr_content(attr, "snippet", &schema->include_snippet); } if (schema->stylesheet) schema->stylesheet_xsp = @@ -240,12 +288,12 @@ static int ioclose_ex(void *context) return 0; } -static void index_field(struct filter_info *tinfo, struct recExtractCtrl *ctrl, +static void index_cdata(struct filter_info *tinfo, struct recExtractCtrl *ctrl, xmlNodePtr ptr, RecWord *recWord) { for(; ptr; ptr = ptr->next) { - index_field(tinfo, ctrl, ptr->children, recWord); + index_cdata(tinfo, ctrl, ptr->children, recWord); if (ptr->type != XML_TEXT_NODE) continue; recWord->term_buf = ptr->content; @@ -261,31 +309,58 @@ static void index_node(struct filter_info *tinfo, struct recExtractCtrl *ctrl, { index_node(tinfo, ctrl, ptr->children, recWord); if (ptr->type != XML_ELEMENT_NODE || !ptr->ns || - strcmp(ptr->ns->href, zebra_index_ns)) + strcmp(ptr->ns->href, zebra_xslt_ns)) continue; if (!strcmp(ptr->name, "index")) { - char *field_str = 0; + const char *name_str = 0; + const char *type_str = 0; const char *xpath_str = 0; struct _xmlAttr *attr; for (attr = ptr->properties; attr; attr = attr->next) { - if (!strcmp(attr->name, "field") - && attr->children && attr->children->type == XML_TEXT_NODE) - field_str = attr->children->content; - if (!strcmp(attr->name, "xpath") - && attr->children && attr->children->type == XML_TEXT_NODE) - xpath_str = attr->children->content; + attr_content(attr, "name", &name_str); + attr_content(attr, "xpath", &xpath_str); + attr_content(attr, "type", &type_str); } - if (field_str) + if (name_str) { - recWord->attrStr = field_str; - index_field(tinfo, ctrl, ptr->children, recWord); + int prev_type = recWord->index_type; /* save default type */ + + if (type_str && *type_str) + recWord->index_type = *type_str; /* type was given */ + recWord->index_name = name_str; + index_cdata(tinfo, ctrl, ptr->children, recWord); + + recWord->index_type = prev_type; /* restore it again */ } } } } +static void index_record(struct filter_info *tinfo,struct recExtractCtrl *ctrl, + xmlNodePtr ptr, RecWord *recWord) +{ + if (ptr->type == XML_ELEMENT_NODE && ptr->ns && + !strcmp(ptr->ns->href, zebra_xslt_ns) + && !strcmp(ptr->name, "record")) + { + const char *type_str = "update"; + const char *id_str = 0; + struct _xmlAttr *attr; + for (attr = ptr->properties; attr; attr = attr->next) + { + attr_content(attr, "type", &type_str); + attr_content(attr, "id", &id_str); + } + if (id_str) + sscanf(id_str, "%255s", ctrl->match_criteria); + + ptr = ptr->children; + } + index_node(tinfo, ctrl, ptr, recWord); +} + static int extract_doc(struct filter_info *tinfo, struct recExtractCtrl *p, xmlDocPtr doc) { @@ -294,13 +369,12 @@ static int extract_doc(struct filter_info *tinfo, struct recExtractCtrl *p, xmlChar *buf_out; int len_out; - struct filter_schema *schema = lookup_schema(tinfo, ZEBRA_INDEX_NS); + struct filter_schema *schema = lookup_schema(tinfo, zebra_xslt_ns); params[0] = 0; - set_param_str(params, "schema", ZEBRA_INDEX_NS, tinfo->odr); + set_param_str(params, "schema", zebra_xslt_ns, tinfo->odr); (*p->init)(p, &recWord); - recWord.reg_type = 'w'; if (schema && schema->stylesheet_xsp) { @@ -313,7 +387,7 @@ static int extract_doc(struct filter_info *tinfo, struct recExtractCtrl *p, fwrite(buf_out, len_out, 1, stdout); xmlFree(buf_out); } - index_node(tinfo, p, xmlDocGetRootElement(resDoc), &recWord); + index_record(tinfo, p, xmlDocGetRootElement(resDoc), &recWord); xmlFreeDoc(resDoc); } xmlDocDumpMemory(doc, &buf_out, &len_out); @@ -412,14 +486,62 @@ static int ioclose_ret(void *context) } +static const char *snippet_doc(struct recRetrieveCtrl *p, int text_mode, + int window_size) +{ + const char *xml_doc_str; + int ord = 0; + WRBUF wrbuf = wrbuf_alloc(); + zebra_snippets *res = + zebra_snippets_window(p->doc_snippet, p->hit_snippet, window_size); + zebra_snippet_word *w = zebra_snippets_list(res); + + if (text_mode) + wrbuf_printf(wrbuf, "\'"); + else + wrbuf_printf(wrbuf, "\n", zebra_xslt_ns); + for (; w; w = w->next) + { + if (ord == 0) + ord = w->ord; + else if (ord != w->ord) + + break; + if (text_mode) + wrbuf_printf(wrbuf, "%s%s%s ", + w->match ? "*" : "", + w->term, + w->match ? "*" : ""); + else + { + wrbuf_printf(wrbuf, " ", + w->ord, w->seqno, + (w->match ? "match='1'" : "")); + wrbuf_xmlputs(wrbuf, w->term); + wrbuf_printf(wrbuf, "\n"); + } + } + if (text_mode) + wrbuf_printf(wrbuf, "\'"); + else + wrbuf_printf(wrbuf, "\n"); + + xml_doc_str = odr_strdup(p->odr, wrbuf_buf(wrbuf)); + + zebra_snippets_destroy(res); + wrbuf_free(wrbuf, 1); + return xml_doc_str; +} + static int filter_retrieve (void *clientData, struct recRetrieveCtrl *p) { - const char *esn = ZEBRA_SCHEMA_IDENTITY_NS; + const char *esn = zebra_xslt_ns; const char *params[10]; struct filter_info *tinfo = clientData; xmlDocPtr resDoc; xmlDocPtr doc; struct filter_schema *schema; + int window_size = -1; if (p->comp) { @@ -439,6 +561,9 @@ static int filter_retrieve (void *clientData, struct recRetrieveCtrl *p) return 0; } + if (schema->include_snippet) + window_size = atoi(schema->include_snippet); + params[0] = 0; set_param_str(params, "schema", esn, p->odr); if (p->fname) @@ -446,7 +571,11 @@ static int filter_retrieve (void *clientData, struct recRetrieveCtrl *p) if (p->score >= 0) set_param_int(params, "score", p->score, p->odr); set_param_int(params, "size", p->recordSize, p->odr); - + set_param_int(params, "id", p->localno, p->odr); + + if (window_size >= 0) + set_param_xml(params, "snippet", snippet_doc(p, 1, window_size), + p->odr); doc = xmlReadIO(ioread_ret, ioclose_ret, p /* I/O handler */, 0 /* URL */, 0 /* encoding */, @@ -457,6 +586,13 @@ static int filter_retrieve (void *clientData, struct recRetrieveCtrl *p) return 0; } + if (window_size >= 0) + { + xmlNodePtr node = xmlDocGetRootElement(doc); + const char *snippet_str = snippet_doc(p, 0, window_size); + xmlDocPtr snippet_doc = xmlParseMemory(snippet_str, strlen(snippet_str)); + xmlAddChild(node, xmlDocGetRootElement(snippet_doc)); + } if (!schema->stylesheet_xsp) resDoc = doc; else