From d9dfda0bf340a35a062c42d566a2390c1f5bc447 Mon Sep 17 00:00:00 2001 From: Adam Dickmeiss Date: Tue, 21 May 2002 07:43:16 +0000 Subject: [PATCH] Comment node. Extra root level for XML parsed data1 --- include/yaz/data1.h | 12 +++++-- retrieval/d1_expat.c | 85 +++++++++++++++++++++++++++++++++++++++-------- retrieval/d1_read.c | 25 +++++++++++--- retrieval/d1_write.c | 90 ++++++++++++++++++++++++++++++-------------------- 4 files changed, 156 insertions(+), 56 deletions(-) diff --git a/include/yaz/data1.h b/include/yaz/data1.h index 430d413..282d301 100644 --- a/include/yaz/data1.h +++ b/include/yaz/data1.h @@ -23,7 +23,7 @@ * LIABILITY, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE * OF THIS SOFTWARE. * - * $Id: data1.h,v 1.10 2002-05-13 14:13:37 adam Exp $ + * $Id: data1.h,v 1.11 2002-05-21 07:43:16 adam Exp $ */ #ifndef DATA1_H @@ -228,7 +228,8 @@ typedef struct data1_node /* variant specification (a triple, actually) */ #define DATA1N_variant 4 int which; - + /* comment (same as data) */ +#define DATA1N_comment 5 union { struct @@ -349,6 +350,13 @@ YAZ_EXPORT data1_node *data1_mk_text_n (data1_handle dh, NMEM mem, YAZ_EXPORT data1_node *data1_mk_text (data1_handle dh, NMEM mem, const char *buf, data1_node *parent); +YAZ_EXPORT data1_node *data1_mk_comment_n (data1_handle dh, NMEM mem, + const char *buf, size_t len, + data1_node *parent); + +YAZ_EXPORT data1_node *data1_mk_comment (data1_handle dh, NMEM mem, + const char *buf, data1_node *parent); + YAZ_EXPORT data1_node *data1_mk_root (data1_handle dh, NMEM nmem, const char *name); diff --git a/retrieval/d1_expat.c b/retrieval/d1_expat.c index 099a922..d308328 100644 --- a/retrieval/d1_expat.c +++ b/retrieval/d1_expat.c @@ -2,7 +2,7 @@ * Copyright (c) 2002, Index Data. * See the file LICENSE for details. * - * $Id: d1_expat.c,v 1.1 2002-05-13 14:13:37 adam Exp $ + * $Id: d1_expat.c,v 1.2 2002-05-21 07:43:16 adam Exp $ */ #if HAVE_EXPAT_H @@ -27,16 +27,10 @@ struct user_info { static void cb_start (void *user, const char *el, const char **attr) { struct user_info *ui = (struct user_info*) user; - if (ui->level) - { - ui->d1_stack[ui->level] = data1_mk_tag (ui->dh, ui->nmem, el, attr, + ui->d1_stack[ui->level] = data1_mk_tag (ui->dh, ui->nmem, el, attr, ui->d1_stack[ui->level-1]); - } - else - { - ui->d1_stack[0] = data1_mk_root (ui->dh, ui->nmem, el); - } ui->level++; + printf ("cb_start %s\n", el); } static void cb_end (void *user, const char *el) @@ -44,13 +38,68 @@ static void cb_end (void *user, const char *el) struct user_info *ui = (struct user_info*) user; ui->level--; + printf ("cb_end %s\n", el); } static void cb_chardata (void *user, const char *s, int len) { struct user_info *ui = (struct user_info*) user; - ui->d1_stack[ui->level] = data1_mk_text_n (ui->dh, ui->nmem, s, len, - ui->d1_stack[ui->level -1]); + int i; + + for (i = 0; id1_stack[ui->level] = data1_mk_text_n (ui->dh, ui->nmem, s, len, + ui->d1_stack[ui->level -1]); + } +} + +static void cb_decl (void *user, const char *version, const char*encoding, + int standalone) +{ + printf ("decl version=%s encoding=%s\n", version ? version : "null", + encoding ? encoding : "null"); +} + +static void cb_processing (void *userData, const char *target, + const char *data) +{ + printf ("decl processing target=%s data=%s\n", target ? target : "null", + data ? data : "null"); +} + +static void cb_comment (void *userData, const char *data) +{ + printf ("decl comment data=%s\n", data ? data : "null"); +} + +static void cb_doctype_start (void *userData, const char *doctypeName, + const char *sysid, const char *pubid, + int has_internal_subset) +{ + printf ("doctype start doctype=%s sysid=%s pubid=%s\n", + doctypeName, sysid, pubid); +} + +static void cb_doctype_end (void *userData) +{ + printf ("doctype end\n"); +} + + +static void cb_entity_decl (void *userData, const char *entityName, + int is_parameter_entity, + const char *value, int value_length, + const char *base, const char *systemId, + const char *publicId, const char *notationName) +{ + printf ("entity %s is_para_entry=%d value=%.*s base=%s systemId=%s\n" + " publicId=%s notationName=%s\n", + entityName, is_parameter_entity, value_length, value, + base, systemId, publicId, notationName); + } #define XML_CHUNK 1024 @@ -63,16 +112,22 @@ data1_node *data1_read_xml (data1_handle dh, struct user_info uinfo; int done = 0; - uinfo.level = 0; + uinfo.level = 1; uinfo.dh = dh; - uinfo.d1_stack[0] = 0; uinfo.nmem = m; - + uinfo.d1_stack[0] = data1_mk_root (dh, m, "root"); + uinfo.d1_stack[1] = 0; /* indicate no children (see end of routine) */ + parser = XML_ParserCreate (0 /* encoding */); XML_SetElementHandler (parser, cb_start, cb_end); XML_SetCharacterDataHandler (parser, cb_chardata); + XML_SetXmlDeclHandler (parser, cb_decl); + XML_SetProcessingInstructionHandler (parser, cb_processing); XML_SetUserData (parser, &uinfo); + XML_SetCommentHandler (parser, cb_comment); + XML_SetDoctypeDeclHandler (parser, cb_doctype_start, cb_doctype_end); + XML_SetEntityDeclHandler (parser, cb_entity_decl); while (!done) { @@ -94,6 +149,8 @@ data1_node *data1_read_xml (data1_handle dh, XML_ParseBuffer (parser, r, done); } XML_ParserFree (parser); + if (!uinfo.d1_stack[1]) + return 0; return uinfo.d1_stack[0]; } diff --git a/retrieval/d1_read.c b/retrieval/d1_read.c index ac19a9f..d2c8863 100644 --- a/retrieval/d1_read.c +++ b/retrieval/d1_read.c @@ -3,7 +3,7 @@ * See the file LICENSE for details. * Sebastian Hammer, Adam Dickmeiss * - * $Id: d1_read.c,v 1.42 2002-05-13 14:13:37 adam Exp $ + * $Id: d1_read.c,v 1.43 2002-05-21 07:43:16 adam Exp $ */ #include @@ -141,8 +141,8 @@ data1_node *data1_mk_tag_n (data1_handle dh, NMEM nmem, while (attr && *attr) { *p = (data1_xattr*) nmem_malloc (nmem, sizeof(**p)); - (*p)->name = nmem_strdup (nmem, attr[0]); - (*p)->value = nmem_strdup (nmem, attr[1]); + (*p)->name = nmem_strdup (nmem, *attr++); + (*p)->value = nmem_strdup (nmem, *attr++); p = &(*p)->next; } *p = 0; @@ -190,13 +190,30 @@ data1_node *data1_mk_text_n (data1_handle dh, NMEM mem, return res; } - data1_node *data1_mk_text (data1_handle dh, NMEM mem, const char *buf, data1_node *parent) { return data1_mk_text_n (dh, mem, buf, strlen(buf), parent); } +data1_node *data1_mk_comment_n (data1_handle dh, NMEM mem, + const char *buf, size_t len, + data1_node *parent) +{ + data1_node *res = data1_mk_node2 (dh, mem, DATA1N_comment, parent); + res->u.data.what = DATA1I_text; + res->u.data.len = len; + + res->u.data.data = data1_insert_string_n (dh, res, mem, buf, len); + return res; +} + +data1_node *data1_mk_comment (data1_handle dh, NMEM mem, + const char *buf, data1_node *parent) +{ + return data1_mk_comment_n (dh, mem, buf, strlen(buf), parent); +} + char *data1_insert_string_n (data1_handle dh, data1_node *res, NMEM m, const char *str, size_t len) { diff --git a/retrieval/d1_write.c b/retrieval/d1_write.c index bf8b8fe..b93a1e7 100644 --- a/retrieval/d1_write.c +++ b/retrieval/d1_write.c @@ -3,7 +3,7 @@ * See the file LICENSE for details. * Sebastian Hammer, Adam Dickmeiss * - * $Id: d1_write.c,v 1.11 2002-05-13 14:13:37 adam Exp $ + * $Id: d1_write.c,v 1.12 2002-05-21 07:43:16 adam Exp $ */ #include @@ -64,50 +64,64 @@ static int nodetoidsgml(data1_node *n, int select, WRBUF b, int col) wrbuf_write(b, line, strlen(line)); } } - else if (c->which == DATA1N_data) + else if (c->which == DATA1N_data || c->which == DATA1N_comment) { char *p = c->u.data.data; int l = c->u.data.len; int first = 1; int lcol = col; - sprintf(line, "%*s", col, ""); - wrbuf_write(b, line, strlen(line)); + if (!c->u.data.formatted_text) + { + sprintf(line, "%*s", col, ""); + wrbuf_write(b, line, strlen(line)); + } + if (c->which == DATA1N_comment) + { + wrbuf_write (b, "", 3); + } } } return 0; -- 1.7.10.4