From 9a6f65b2a18219811d38b6e4ba8a2ddb567dca53 Mon Sep 17 00:00:00 2001 From: Lev Walkin Date: Sat, 23 Oct 2004 11:20:06 +0000 Subject: [PATCH] XER decoding support --- skeletons/BMPString.c | 77 +++++++++++++++++++++++++++-- skeletons/BMPString.h | 1 + skeletons/UTF8String.c | 96 ++++++++++++++++++++++--------------- skeletons/UTF8String.h | 13 +++++ skeletons/UniversalString.c | 68 ++++++++++++++++++++++++-- skeletons/UniversalString.h | 1 + skeletons/file-dependencies | 4 +- 7 files changed, 209 insertions(+), 51 deletions(-) diff --git a/skeletons/BMPString.c b/skeletons/BMPString.c index cca6e3c5..3dd01d55 100644 --- a/skeletons/BMPString.c +++ b/skeletons/BMPString.c @@ -1,9 +1,11 @@ /*- - * Copyright (c) 2003 Lev Walkin . All rights reserved. + * Copyright (c) 2003, 2004 Lev Walkin . All rights reserved. * Redistribution and modifications are permitted subject to BSD license. */ #include #include +#include +#include /* * BMPString basic type description. @@ -18,10 +20,10 @@ asn_TYPE_descriptor_t asn_DEF_BMPString = { OCTET_STRING_free, /* Implemented in terms of OCTET STRING */ BMPString_print, asn_generic_no_constraint, /* No constraint by default */ - OCTET_STRING_decode_ber, /* Implemented in terms of OCTET STRING */ - OCTET_STRING_encode_der, /* Implemented in terms of OCTET STRING */ - 0, /* Not implemented yet */ - BMPString_encode_xer, /* Convert to UTF8 */ + OCTET_STRING_decode_ber, + OCTET_STRING_encode_der, + BMPString_decode_xer, /* Convert from UTF-8 */ + BMPString_encode_xer, /* Convert to UTF-8 */ 0, /* Use generic outmost tag fetcher */ asn_DEF_BMPString_tags, sizeof(asn_DEF_BMPString_tags) @@ -74,6 +76,71 @@ BMPString__dump(const BMPString_t *st, return wrote; } +asn_dec_rval_t +BMPString_decode_xer(asn_codec_ctx_t *opt_codec_ctx, + asn_TYPE_descriptor_t *td, void **sptr, + const char *opt_mname, void *buf_ptr, size_t size) { + asn_dec_rval_t rc; + + rc = OCTET_STRING_decode_xer_utf8(opt_codec_ctx, td, sptr, opt_mname, + buf_ptr, size); + if(rc.code == RC_OK) { + /* + * Now we have a whole string in UTF-8 format. + * Convert it into UCS-2. + */ + uint32_t *wcs; + size_t wcs_len; + UTF8String_t *st; + + assert(*sptr); + st = (UTF8String_t *)*sptr; + assert(st->buf); + wcs_len = UTF8String_to_wcs(st, 0, 0); + + wcs = (uint32_t *)MALLOC(4 * (wcs_len + 1)); + if(wcs == 0 || UTF8String_to_wcs(st, wcs, wcs_len) != wcs_len) { + rc.code = RC_FAIL; + rc.consumed = 0; + return rc; + } else { + wcs[wcs_len] = 0; /* nul-terminate */ + } + + if(1) { + /* Swap byte order and trim encoding to 2 bytes */ + uint32_t *wc = wcs; + uint32_t *wc_end = wcs + wcs_len + 1; + uint16_t *dstwc = (uint16_t *)wcs; + for(; wc < wc_end; wc++, dstwc++) { + uint32_t wch = *wc; + if(wch > 0xffff) { + FREEMEM(wcs); + rc.code = RC_FAIL; + rc.consumed = 0; + return rc; + } + *((uint8_t *)dstwc + 0) = wch >> 8; + *((uint8_t *)dstwc + 1) = wch; + } + dstwc = (uint16_t)REALLOC(wcs, 2 * (wcs_len + 1)); + if(!dstwc) { + FREEMEM(wcs); + rc.code = RC_FAIL; + rc.consumed = 0; + return rc; + } else { + wcs = (uint32_t *)dstwc; + } + } + + FREEMEM(st->buf); + st->buf = (uint8_t *)wcs; + st->size = 2 * wcs_len; + } + return rc; +} + asn_enc_rval_t BMPString_encode_xer(asn_TYPE_descriptor_t *td, void *sptr, int ilevel, enum xer_encoder_flags_e flags, diff --git a/skeletons/BMPString.h b/skeletons/BMPString.h index 3ff2baf3..cd515973 100644 --- a/skeletons/BMPString.h +++ b/skeletons/BMPString.h @@ -12,6 +12,7 @@ typedef OCTET_STRING_t BMPString_t; /* Implemented via OCTET STRING */ extern asn_TYPE_descriptor_t asn_DEF_BMPString; asn_struct_print_f BMPString_print; /* Human-readable output */ +xer_type_decoder_f BMPString_decode_xer; xer_type_encoder_f BMPString_encode_xer; #endif /* _BMPString_H_ */ diff --git a/skeletons/UTF8String.c b/skeletons/UTF8String.c index 1c0c731d..0a47f459 100644 --- a/skeletons/UTF8String.c +++ b/skeletons/UTF8String.c @@ -96,54 +96,72 @@ UTF8String_constraint(asn_TYPE_descriptor_t *td, const void *sptr, return (len < 0) ? -1 : 0; } -ssize_t -UTF8String_length(const UTF8String_t *st) { +static ssize_t +UTF8String__process(const UTF8String_t *st, uint32_t *dst, size_t dstlen) { + size_t length; + uint8_t *buf = st->buf; + uint8_t *end = buf + st->size; - if(st && st->buf) { - size_t length; - uint8_t *buf = st->buf; - uint8_t *end = buf + st->size; + for(length = 0; buf < end; length++) { + int ch = *buf; + uint8_t *cend; + int32_t value; + int want; - for(length = 0; buf < end; length++) { - int ch = *buf; - uint8_t *cend; - int32_t value; - int want; - - /* Compute the sequence length */ - want = UTF8String_ht[0][ch >> 4]; - switch(want) { - case -1: - /* Second half of the table, long sequence */ - want = UTF8String_ht[1][ch & 0x0F]; - if(want != -1) break; - /* Fall through */ - case 0: - return U8E_ILLSTART; - } - - /* assert(want >= 1 && want <= 6) */ - - /* Check character sequence length */ - if(buf + want > end) return U8E_TRUNC; - - value = ch & (0xff >> (want + 1)); - cend = buf + want; - for(buf++; buf < cend; buf++) { - ch = *buf; - if(ch < 0x80 || ch > 0xbf) return U8E_NOTCONT; - value = (value << 6) | (ch & 0x3F); - } - if(value < UTF8String_mv[want]) - return U8E_NOTMIN; + /* Compute the sequence length */ + want = UTF8String_ht[0][ch >> 4]; + switch(want) { + case -1: + /* Second half of the table, long sequence */ + want = UTF8String_ht[1][ch & 0x0F]; + if(want != -1) break; + /* Fall through */ + case 0: + return U8E_ILLSTART; } - return length; + /* assert(want >= 1 && want <= 6) */ + + /* Check character sequence length */ + if(buf + want > end) return U8E_TRUNC; + + value = ch & (0xff >> (want + 1)); + cend = buf + want; + for(buf++; buf < cend; buf++) { + ch = *buf; + if(ch < 0x80 || ch > 0xbf) return U8E_NOTCONT; + value = (value << 6) | (ch & 0x3F); + } + if(value < UTF8String_mv[want]) + return U8E_NOTMIN; + if(dstlen) *dst++ = value; /* Record value */ + } + + if(dstlen) *dst = 0; /* zero-terminate */ + + return length; +} + + +ssize_t +UTF8String_length(const UTF8String_t *st) { + if(st && st->buf) { + return UTF8String__process(st, 0, 0); } else { return U8E_EINVAL; } } +size_t +UTF8String_to_wcs(const UTF8String_t *st, uint32_t *dst, size_t dstlen) { + if(st && st->buf) { + ssize_t ret = UTF8String__process(st, dst, dstlen); + return (ret < 0) ? 0 : ret; + } else { + return 0; + } +} + int UTF8String_print(asn_TYPE_descriptor_t *td, const void *sptr, int ilevel, asn_app_consume_bytes_f *cb, void *app_key) { diff --git a/skeletons/UTF8String.h b/skeletons/UTF8String.h index 45801592..84e6eaff 100644 --- a/skeletons/UTF8String.h +++ b/skeletons/UTF8String.h @@ -25,4 +25,17 @@ asn_constr_check_f UTF8String_constraint; */ ssize_t UTF8String_length(const UTF8String_t *st); +/* + * Convert the UTF-8 string into a sequence of wide characters. + * Returns the number of characters necessary. + * Returned value might be greater than dstlen. + * In case of conversion error, 0 is returned. + * + * If st points to a valid UTF-8 string, calling + * UTF8String_to_wcs(st, 0, 0); + * is equivalent to + * UTF8String_length(const UTF8String_t *st); + */ +size_t UTF8String_to_wcs(const UTF8String_t *st, uint32_t *dst, size_t dstlen); + #endif /* _UTF8String_H_ */ diff --git a/skeletons/UniversalString.c b/skeletons/UniversalString.c index 2221000f..f3279b40 100644 --- a/skeletons/UniversalString.c +++ b/skeletons/UniversalString.c @@ -1,9 +1,10 @@ /*- - * Copyright (c) 2003 Lev Walkin . All rights reserved. + * Copyright (c) 2003, 2004 Lev Walkin . All rights reserved. * Redistribution and modifications are permitted subject to BSD license. */ #include #include +#include /* * UniversalString basic type description. @@ -18,10 +19,10 @@ asn_TYPE_descriptor_t asn_DEF_UniversalString = { OCTET_STRING_free, UniversalString_print, /* Convert into UTF8 and print */ asn_generic_no_constraint, - OCTET_STRING_decode_ber, /* Implemented in terms of OCTET STRING */ - OCTET_STRING_encode_der, /* Implemented in terms of OCTET STRING */ - 0, /* Not implemented yet */ - UniversalString_encode_xer, /* Convert into UTF8 */ + OCTET_STRING_decode_ber, + OCTET_STRING_encode_der, + UniversalString_decode_xer, /* Convert from UTF-8 */ + UniversalString_encode_xer, /* Convert into UTF-8 */ 0, /* Use generic outmost tag fetcher */ asn_DEF_UniversalString_tags, sizeof(asn_DEF_UniversalString_tags) @@ -93,6 +94,63 @@ UniversalString__dump(const UniversalString_t *st, return wrote; } +asn_dec_rval_t +UniversalString_decode_xer(asn_codec_ctx_t *opt_codec_ctx, + asn_TYPE_descriptor_t *td, void **sptr, + const char *opt_mname, void *buf_ptr, size_t size) { + asn_dec_rval_t rc; + + rc = OCTET_STRING_decode_xer_utf8(opt_codec_ctx, td, sptr, opt_mname, + buf_ptr, size); + if(rc.code == RC_OK) { + /* + * Now we have a whole string in UTF-8 format. + * Convert it into UCS-4. + */ + uint32_t *wcs; + size_t wcs_len; + UTF8String_t *st; +#ifndef WORDS_BIGENDIAN + int little_endian = 1; +#endif + + assert(*sptr); + st = (UTF8String_t *)*sptr; + assert(st->buf); + wcs_len = UTF8String_to_wcs(st, 0, 0); + + wcs = (uint32_t *)MALLOC(4 * (wcs_len + 1)); + if(wcs == 0 || UTF8String_to_wcs(st, wcs, wcs_len) != wcs_len) { + rc.code = RC_FAIL; + rc.consumed = 0; + return rc; + } else { + wcs[wcs_len] = 0; /* nul-terminate */ + } + +#ifndef WORDS_BIGENDIAN + if(*(char *)&little_endian) { + /* Swap byte order in encoding */ + uint32_t *wc = wcs; + uint32_t *wc_end = wcs + wcs_len; + for(; wc < wc_end; wc++) { + /* *wc = htonl(*wc); */ + uint32_t wch = *wc; + *((uint8_t *)wc + 0) = wch >> 24; + *((uint8_t *)wc + 1) = wch >> 16; + *((uint8_t *)wc + 2) = wch >> 8; + *((uint8_t *)wc + 3) = wch; + } + } +#endif /* WORDS_BIGENDIAN */ + + FREEMEM(st->buf); + st->buf = (uint8_t *)wcs; + st->size = 4 * wcs_len; + } + return rc; +} + asn_enc_rval_t UniversalString_encode_xer(asn_TYPE_descriptor_t *td, void *sptr, int ilevel, enum xer_encoder_flags_e flags, diff --git a/skeletons/UniversalString.h b/skeletons/UniversalString.h index 6d4a291f..3be8cabd 100644 --- a/skeletons/UniversalString.h +++ b/skeletons/UniversalString.h @@ -12,6 +12,7 @@ typedef OCTET_STRING_t UniversalString_t; /* Implemented via OCTET STRING */ extern asn_TYPE_descriptor_t asn_DEF_UniversalString; asn_struct_print_f UniversalString_print; /* Human-readable output */ +xer_type_decoder_f UniversalString_decode_xer; xer_type_encoder_f UniversalString_encode_xer; #endif /* _UniversalString_H_ */ diff --git a/skeletons/file-dependencies b/skeletons/file-dependencies index 3e26fffb..3c8715fa 100644 --- a/skeletons/file-dependencies +++ b/skeletons/file-dependencies @@ -7,7 +7,7 @@ # ANY.h ANY.c -BMPString.h BMPString.c +BMPString.h BMPString.c UTF8String.h BOOLEAN.h BOOLEAN.c ENUMERATED.h ENUMERATED.c INTEGER.h GeneralString.h GeneralString.c @@ -30,7 +30,7 @@ T61String.h T61String.c TeletexString.h TeletexString.c UTCTime.h UTCTime.c GeneralizedTime.h UTF8String.h UTF8String.c -UniversalString.h UniversalString.c +UniversalString.h UniversalString.c UTF8String.h VideotexString.h VideotexString.c VisibleString.h VisibleString.c asn_SEQUENCE_OF.h asn_SEQUENCE_OF.c asn_SET_OF.h