diff options
Diffstat (limited to 'src/lib/pair.c')
-rw-r--r-- | src/lib/pair.c | 2518 |
1 files changed, 2518 insertions, 0 deletions
diff --git a/src/lib/pair.c b/src/lib/pair.c new file mode 100644 index 0000000..898e1e8 --- /dev/null +++ b/src/lib/pair.c @@ -0,0 +1,2518 @@ +/* + * pair.c Functions to handle VALUE_PAIRs + * + * Version: $Id$ + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA + * + * Copyright 2000,2006 The FreeRADIUS server project + */ + +RCSID("$Id$") + +#include <freeradius-devel/libradius.h> +#include <freeradius-devel/regex.h> + +#include <ctype.h> + +/** Free a VALUE_PAIR + * + * @note Do not call directly, use talloc_free instead. + * + * @param vp to free. + * @return 0 + */ +static int _fr_pair_free(VALUE_PAIR *vp) { +#ifndef NDEBUG + vp->vp_integer = 0xf4eef4ee; +#endif + +#ifdef TALLOC_DEBUG + talloc_report_depth_cb(NULL, 0, -1, fr_talloc_verify_cb, NULL); +#endif + return 0; +} + +VALUE_PAIR *fr_pair_alloc(TALLOC_CTX *ctx) +{ + VALUE_PAIR *vp; + + vp = talloc_zero(ctx, VALUE_PAIR); + if (!vp) { + fr_strerror_printf("Out of memory"); + return NULL; + } + + vp->op = T_OP_EQ; + vp->tag = TAG_ANY; + vp->type = VT_NONE; + + talloc_set_destructor(vp, _fr_pair_free); + + return vp; +} + + +/** Dynamically allocate a new attribute + * + * Allocates a new attribute and a new dictionary attr if no DA is provided. + * + * @param[in] ctx for allocated memory, usually a pointer to a RADIUS_PACKET + * @param[in] da Specifies the dictionary attribute to build the VP from. + * @return a new value pair or NULL if an error occurred. + */ +VALUE_PAIR *fr_pair_afrom_da(TALLOC_CTX *ctx, DICT_ATTR const *da) +{ + VALUE_PAIR *vp; + + /* + * Caller must specify a da else we don't know what the attribute type is. + */ + if (!da) { + fr_strerror_printf("Invalid arguments"); + return NULL; + } + + vp = fr_pair_alloc(ctx); + if (!vp) { + fr_strerror_printf("Out of memory"); + return NULL; + } + + /* + * Use the 'da' to initialize more fields. + */ + vp->da = da; + vp->vp_length = da->flags.length; + + return vp; +} + +/** Create a new valuepair + * + * If attr and vendor match a dictionary entry then a VP with that DICT_ATTR + * will be returned. + * + * If attr or vendor are uknown will call dict_attruknown to create a dynamic + * DICT_ATTR of PW_TYPE_OCTETS. + * + * Which type of DICT_ATTR the VALUE_PAIR was created with can be determined by + * checking @verbatim vp->da->flags.is_unknown @endverbatim. + * + * @param[in] ctx for allocated memory, usually a pointer to a RADIUS_PACKET + * @param[in] attr number. + * @param[in] vendor number. + * @return the new valuepair or NULL on error. + */ +VALUE_PAIR *fr_pair_afrom_num(TALLOC_CTX *ctx, unsigned int attr, unsigned int vendor) +{ + DICT_ATTR const *da; + + da = dict_attrbyvalue(attr, vendor); + if (!da) return NULL; + + return fr_pair_afrom_da(ctx, da); +} + +/** Free memory used by a valuepair list. + * + * @todo TLV: needs to free all dependents of each VP freed. + */ +void fr_pair_list_free(VALUE_PAIR **vps) +{ + VALUE_PAIR *vp; + vp_cursor_t cursor; + + if (!vps || !*vps) { + return; + } + + for (vp = fr_cursor_init(&cursor, vps); + vp; + vp = fr_cursor_next(&cursor)) { + VERIFY_VP(vp); + talloc_free(vp); + } + + *vps = NULL; +} + +/** Mark malformed or unrecognised attributed as unknown + * + * @param vp to change DICT_ATTR of. + * @return 0 on success (or if already unknown) else -1 on error. + */ +int fr_pair_to_unknown(VALUE_PAIR *vp) +{ + DICT_ATTR const *da; + + VERIFY_VP(vp); + if (vp->da->flags.is_unknown) { + return 0; + } + + da = dict_unknown_afrom_fields(vp, vp->da->attr, vp->da->vendor); + if (!da) { + return -1; + } + + vp->da = da; + + return 0; +} + +/** Find the pair with the matching DAs + * + */ +VALUE_PAIR *fr_pair_find_by_da(VALUE_PAIR *vp, DICT_ATTR const *da, int8_t tag) +{ + vp_cursor_t cursor; + + if(!fr_assert(da)) { + return NULL; + } + + (void) fr_cursor_init(&cursor, &vp); + return fr_cursor_next_by_da(&cursor, da, tag); +} + + +/** Find the pair with the matching attribute + * + * @todo should take DAs and do a pointer comparison. + */ +VALUE_PAIR *fr_pair_find_by_num(VALUE_PAIR *vp, unsigned int attr, unsigned int vendor, int8_t tag) +{ + vp_cursor_t cursor; + + /* List head may be NULL if it contains no VPs */ + if (!vp) return NULL; + + VERIFY_LIST(vp, ""); + + (void) fr_cursor_init(&cursor, &vp); + return fr_cursor_next_by_num(&cursor, attr, vendor, tag); +} + +/** Delete matching pairs + * + * Delete matching pairs from the attribute list. + * + * @param[in,out] first VP in list. + * @param[in] attr to match. + * @param[in] vendor to match. + * @param[in] tag to match. TAG_ANY matches any tag, TAG_NONE matches tagless VPs. + * + * @todo should take DAs and do a point comparison. + */ +void fr_pair_delete_by_num(VALUE_PAIR **first, unsigned int attr, unsigned int vendor, int8_t tag) +{ + VALUE_PAIR *i, *next; + VALUE_PAIR **last = first; + + for(i = *first; i; i = next) { + VERIFY_VP(i); + next = i->next; + if ((i->da->attr == attr) && (i->da->vendor == vendor) && + (!i->da->flags.has_tag || TAG_EQ(tag, i->tag))) { + *last = next; + talloc_free(i); + } else { + last = &i->next; + } + } +} + +/** Delete matching pairs by da + * + * Delete matching pairs from the attribute list. + * + * @param[in,out] first VP in list. + * @param[in] da to match. + */ +void fr_pair_delete_by_da(VALUE_PAIR **first, DICT_ATTR const *da) +{ + VALUE_PAIR *i, *next; + VALUE_PAIR **last = first; + + for(i = *first; i; i = next) { + VERIFY_VP(i); + next = i->next; + if (i->da == da) { + *last = next; + talloc_free(i); + } else { + last = &i->next; + } + } +} + +/** Add a VP to the end of the list. + * + * Locates the end of 'first', and links an additional VP 'add' at the end. + * + * @param[in] first VP in linked list. Will add new VP to the end of this list. + * @param[in] add VP to add to list. + */ +void fr_pair_add(VALUE_PAIR **first, VALUE_PAIR *add) +{ + VALUE_PAIR *i; + + if (!add) return; + + VERIFY_VP(add); + + if (*first == NULL) { + *first = add; + return; + } + + for (i = *first; i->next; i = i->next) { +#ifdef WITH_VERIFY_PTR + VERIFY_VP(i); + /* + * The same VP should never by added multiple times + * to the same list. + */ + fr_assert(i != add); +#endif + } + + i->next = add; +} + +/** Add a VP to the start of the list. + * + * Links the new VP to 'first', then points 'first' at the new VP. + * + * @param[in] first VP in linked list. Will add new VP to the start of this list. + * @param[in] add VP to add to list. + */ +void fr_pair_prepend(VALUE_PAIR **first, VALUE_PAIR *add) +{ + VALUE_PAIR *i; + + if (!add) return; + + VERIFY_VP(add); + + if (*first == NULL) { + *first = add; + return; + } + + /* + * Find the end of the list to be prepended + */ + for (i = add; i->next; i = i->next) { +#ifdef WITH_VERIFY_PTR + VERIFY_VP(i); + /* + * The same VP should never by added multiple times + * to the same list. + */ + fr_assert(*first != i); +#endif + } + i->next = *first; + *first = add; +} + +/** Replace all matching VPs + * + * Walks over 'first', and replaces the first VP that matches 'replace'. + * + * @note Memory used by the VP being replaced will be freed. + * @note Will not work with unknown attributes. + * + * @param[in,out] first VP in linked list. Will search and replace in this list. + * @param[in] replace VP to replace. + */ +void fr_pair_replace(VALUE_PAIR **first, VALUE_PAIR *replace) +{ + VALUE_PAIR *i, *next; + VALUE_PAIR **prev = first; + + VERIFY_VP(replace); + + if (*first == NULL) { + *first = replace; + return; + } + + /* + * Not an empty list, so find item if it is there, and + * replace it. Note, we always replace the first one, and + * we ignore any others that might exist. + */ + for(i = *first; i; i = next) { + VERIFY_VP(i); + next = i->next; + + /* + * Found the first attribute, replace it, + * and return. + */ + if ((i->da == replace->da) && (!i->da->flags.has_tag || TAG_EQ(replace->tag, i->tag))) { + *prev = replace; + + /* + * Should really assert that replace->next == NULL + */ + replace->next = next; + talloc_free(i); + return; + } + + /* + * Point to where the attribute should go. + */ + prev = &i->next; + } + + /* + * If we got here, we didn't find anything to replace, so + * stopped at the last item, which we just append to. + */ + *prev = replace; +} + +int8_t fr_pair_cmp_by_da_tag(void const *a, void const *b) +{ + VALUE_PAIR const *my_a = a; + VALUE_PAIR const *my_b = b; + + VERIFY_VP(my_a); + VERIFY_VP(my_b); + + uint8_t cmp; + + cmp = fr_pointer_cmp(my_a->da, my_b->da); + if (cmp != 0) return cmp; + + if (my_a->tag < my_b->tag) return -1; + + if (my_a->tag > my_b->tag) return 1; + + return 0; +} + +static void fr_pair_list_sort_split(VALUE_PAIR *source, VALUE_PAIR **front, VALUE_PAIR **back) +{ + VALUE_PAIR *fast; + VALUE_PAIR *slow; + + /* + * Stopping condition - no more elements left to split + */ + if (!source || !source->next) { + *front = source; + *back = NULL; + + return; + } + + /* + * Fast advances twice as fast as slow, so when it gets to the end, + * slow will point to the middle of the linked list. + */ + slow = source; + fast = source->next; + + while (fast) { + fast = fast->next; + if (fast) { + slow = slow->next; + fast = fast->next; + } + } + + *front = source; + *back = slow->next; + slow->next = NULL; +} + +static VALUE_PAIR *fr_pair_list_sort_merge(VALUE_PAIR *a, VALUE_PAIR *b, fr_cmp_t cmp) +{ + VALUE_PAIR *result = NULL; + + if (!a) return b; + if (!b) return a; + + /* + * Compare the DICT_ATTRs and tags + */ + if (cmp(a, b) <= 0) { + result = a; + result->next = fr_pair_list_sort_merge(a->next, b, cmp); + } else { + result = b; + result->next = fr_pair_list_sort_merge(a, b->next, cmp); + } + + return result; +} + +/** Sort a linked list of VALUE_PAIRs using merge sort + * + * @param[in,out] vps List of VALUE_PAIRs to sort. + * @param[in] cmp to sort with + */ +void fr_pair_list_sort(VALUE_PAIR **vps, fr_cmp_t cmp) +{ + VALUE_PAIR *head = *vps; + VALUE_PAIR *a; + VALUE_PAIR *b; + + /* + * If there's 0-1 elements it must already be sorted. + */ + if (!head || !head->next) { + return; + } + + fr_pair_list_sort_split(head, &a, &b); /* Split into sublists */ + fr_pair_list_sort(&a, cmp); /* Traverse left */ + fr_pair_list_sort(&b, cmp); /* Traverse right */ + + /* + * merge the two sorted lists together + */ + *vps = fr_pair_list_sort_merge(a, b, cmp); +} + +/** Write an error to the library errorbuff detailing the mismatch + * + * Retrieve output with fr_strerror(); + * + * @todo add thread specific talloc contexts. + * + * @param ctx a hack until we have thread specific talloc contexts. + * @param failed pair of attributes which didn't match. + */ +void fr_pair_validate_debug(TALLOC_CTX *ctx, VALUE_PAIR const *failed[2]) +{ + VALUE_PAIR const *filter = failed[0]; + VALUE_PAIR const *list = failed[1]; + + char *value, *str; + + (void) fr_strerror(); /* Clear any existing messages */ + + if (!fr_assert(!(!filter && !list))) return; + + if (!list) { + if (!filter) return; + fr_strerror_printf("Attribute \"%s\" not found in list", filter->da->name); + return; + } + + if (!filter || (filter->da != list->da)) { + fr_strerror_printf("Attribute \"%s\" not found in filter", list->da->name); + return; + } + + if (!TAG_EQ(filter->tag, list->tag)) { + fr_strerror_printf("Attribute \"%s\" tag \"%i\" didn't match filter tag \"%i\"", + list->da->name, list->tag, filter->tag); + return; + } + + + value = vp_aprints_value(ctx, list, '"'); + str = vp_aprints(ctx, filter, '"'); + + fr_strerror_printf("Attribute value \"%s\" didn't match filter: %s", value, str); + + talloc_free(str); + talloc_free(value); + + return; +} + +/** Uses fr_pair_cmp to verify all VALUE_PAIRs in list match the filter defined by check + * + * @note will sort both filter and list in place. + * + * @param failed pointer to an array to write the pointers of the filter/list attributes that didn't match. + * May be NULL. + * @param filter attributes to check list against. + * @param list attributes, probably a request or reply + */ +bool fr_pair_validate(VALUE_PAIR const *failed[2], VALUE_PAIR *filter, VALUE_PAIR *list) +{ + vp_cursor_t filter_cursor; + vp_cursor_t list_cursor; + + VALUE_PAIR *check, *match; + + if (!filter && !list) { + return true; + } + + /* + * This allows us to verify the sets of validate and reply are equal + * i.e. we have a validate rule which matches every reply attribute. + * + * @todo this should be removed one we have sets and lists + */ + fr_pair_list_sort(&filter, fr_pair_cmp_by_da_tag); + fr_pair_list_sort(&list, fr_pair_cmp_by_da_tag); + + check = fr_cursor_init(&filter_cursor, &filter); + match = fr_cursor_init(&list_cursor, &list); + while (match || check) { + /* + * Lists are of different lengths + */ + if (!match || !check) goto mismatch; + + /* + * The lists are sorted, so if the first + * attributes aren't of the same type, then we're + * done. + */ + if (!ATTRIBUTE_EQ(check, match)) goto mismatch; + + /* + * They're of the same type, but don't have the + * same values. This is a problem. + * + * Note that the RFCs say that for attributes of + * the same type, order is important. + */ + if (fr_pair_cmp(check, match) != 1) goto mismatch; + + check = fr_cursor_next(&filter_cursor); + match = fr_cursor_next(&list_cursor); + } + + return true; + +mismatch: + if (failed) { + failed[0] = check; + failed[1] = match; + } + return false; +} + +/** Uses fr_pair_cmp to verify all VALUE_PAIRs in list match the filter defined by check + * + * @note will sort both filter and list in place. + * + * @param failed pointer to an array to write the pointers of the filter/list attributes that didn't match. + * May be NULL. + * @param filter attributes to check list against. + * @param list attributes, probably a request or reply + */ +bool fr_pair_validate_relaxed(VALUE_PAIR const *failed[2], VALUE_PAIR *filter, VALUE_PAIR *list) +{ + vp_cursor_t filter_cursor; + vp_cursor_t list_cursor; + + VALUE_PAIR *check, *last_check = NULL, *match = NULL; + + if (!filter && !list) { + return true; + } + + /* + * This allows us to verify the sets of validate and reply are equal + * i.e. we have a validate rule which matches every reply attribute. + * + * @todo this should be removed one we have sets and lists + */ + fr_pair_list_sort(&filter, fr_pair_cmp_by_da_tag); + fr_pair_list_sort(&list, fr_pair_cmp_by_da_tag); + + fr_cursor_init(&list_cursor, &list); + for (check = fr_cursor_init(&filter_cursor, &filter); + check; + check = fr_cursor_next(&filter_cursor)) { + /* + * Were processing check attributes of a new type. + */ + if (!ATTRIBUTE_EQ(last_check, check)) { + /* + * Record the start of the matching attributes in the pair list + * For every other operator we require the match to be present + */ + match = fr_cursor_next_by_da(&list_cursor, check->da, check->tag); + if (!match) { + if (check->op == T_OP_CMP_FALSE) continue; + goto mismatch; + } + + fr_cursor_init(&list_cursor, &match); + last_check = check; + } + + /* + * Now iterate over all attributes of the same type. + */ + for (match = fr_cursor_first(&list_cursor); + ATTRIBUTE_EQ(match, check); + match = fr_cursor_next(&list_cursor)) { + /* + * This attribute passed the filter + */ + if (!fr_pair_cmp(check, match)) goto mismatch; + } + } + + return true; + +mismatch: + if (failed) { + failed[0] = check; + failed[1] = match; + } + return false; +} + +/** Copy a single valuepair + * + * Allocate a new valuepair and copy the da from the old vp. + * + * @param[in] ctx for talloc + * @param[in] vp to copy. + * @return a copy of the input VP or NULL on error. + */ +VALUE_PAIR *fr_pair_copy(TALLOC_CTX *ctx, VALUE_PAIR const *vp) +{ + VALUE_PAIR *n; + + if (!vp) return NULL; + + VERIFY_VP(vp); + + n = fr_pair_afrom_da(ctx, vp->da); + if (!n) return NULL; + + memcpy(n, vp, sizeof(*n)); + + /* + * If the DA is unknown, steal "n" to "ctx". This does + * nothing for "n", but will also copy the unknown "da". + */ + if (n->da->flags.is_unknown) { + fr_pair_steal(ctx, n); + } + + n->next = NULL; + + /* + * If it's an xlat, copy the raw string and return early, + * so we don't pre-expand or otherwise mangle the VALUE_PAIR. + */ + if (vp->type == VT_XLAT) { + n->value.xlat = talloc_typed_strdup(n, n->value.xlat); + return n; + } + + switch (vp->da->type) { + case PW_TYPE_OCTETS: + n->vp_octets = NULL; /* else fr_pair_value_memcpy will free vp's value */ + fr_pair_value_memcpy(n, vp->vp_octets, n->vp_length); + break; + + case PW_TYPE_STRING: + n->vp_strvalue = NULL; /* else pairstrnpy will free vp's value */ + fr_pair_value_bstrncpy(n, vp->vp_strvalue, n->vp_length); + break; + + default: + break; + } + + return n; +} + +/** Copy a pairlist. + * + * Copy all pairs from 'from' regardless of tag, attribute or vendor. + * + * @param[in] ctx for new VALUE_PAIRs to be allocated in. + * @param[in] from whence to copy VALUE_PAIRs. + * @return the head of the new VALUE_PAIR list or NULL on error. + */ +VALUE_PAIR *fr_pair_list_copy(TALLOC_CTX *ctx, VALUE_PAIR *from) +{ + vp_cursor_t src, dst; + + VALUE_PAIR *out = NULL, *vp; + + fr_cursor_init(&dst, &out); + for (vp = fr_cursor_init(&src, &from); + vp; + vp = fr_cursor_next(&src)) { + VERIFY_VP(vp); + vp = fr_pair_copy(ctx, vp); + if (!vp) { + fr_pair_list_free(&out); + return NULL; + } + fr_cursor_insert(&dst, vp); /* fr_pair_list_copy sets next pointer to NULL */ + } + + return out; +} + +/** Copy matching pairs + * + * Copy pairs of a matching attribute number, vendor number and tag from the + * the input list to a new list, and returns the head of this list. + * + * @param[in] ctx for talloc + * @param[in] from whence to copy VALUE_PAIRs. + * @param[in] attr to match. If attribute PW_VENDOR_SPECIFIC and vendor 0, + * will match (and therefore copy) only VSAs. + * If attribute 0 and vendor 0 will match (and therefore copy) all + * attributes. + * @param[in] vendor to match. + * @param[in] tag to match, TAG_ANY matches any tag, TAG_NONE matches tagless VPs. + * @return the head of the new VALUE_PAIR list or NULL on error. + */ +VALUE_PAIR *fr_pair_list_copy_by_num(TALLOC_CTX *ctx, VALUE_PAIR *from, + unsigned int attr, unsigned int vendor, int8_t tag) +{ + vp_cursor_t src, dst; + + VALUE_PAIR *out = NULL, *vp; + + fr_cursor_init(&dst, &out); + for (vp = fr_cursor_init(&src, &from); + vp; + vp = fr_cursor_next(&src)) { + VERIFY_VP(vp); + + if (vp->da->flags.has_tag && !TAG_EQ(tag, vp->tag)) { + continue; + } + + /* + * Attr/vendor of 0 means "move them all". + * It's better than "fr_pair_copy(foo,bar);bar=NULL" + */ + if ((attr == 0) && (vendor == 0)) { + goto do_copy; + } + + /* + * vendor=0, attr = PW_VENDOR_SPECIFIC means + * "match any vendor attribute". + */ + if ((vendor == 0) && (attr == PW_VENDOR_SPECIFIC)) { + /* + * It's a VSA: copy it over. + */ + if (vp->da->vendor != 0) goto do_copy; + + /* + * It's Vendor-Specific: copy it over. + */ + if (vp->da->attr == attr) goto do_copy; + + /* + * It's not a VSA: ignore it. + */ + continue; + } + + if ((vp->da->attr != attr) || (vp->da->vendor != vendor)) { + continue; + } + + do_copy: + vp = fr_pair_copy(ctx, vp); + if (!vp) { + fr_pair_list_free(&out); + return NULL; + } + fr_cursor_insert(&dst, vp); + } + + return out; +} + +/** Steal one VP + * + * @param[in] ctx to move VALUE_PAIR into + * @param[in] vp VALUE_PAIR to move into the new context. + */ +void fr_pair_steal(TALLOC_CTX *ctx, VALUE_PAIR *vp) +{ + (void) talloc_steal(ctx, vp); + + /* + * The DA may be unknown. If we're stealing the VPs to a + * different context, copy the unknown DA. We use the VP + * as a context here instead of "ctx", so that when the + * VP is freed, so is the DA. + * + * Since we have no introspection into OTHER VPs using + * the same DA, we can't have multiple VPs use the same + * DA. So we might as well tie it to this VP. + */ + if (vp->da->flags.is_unknown) { + DICT_ATTR *da; + char *p; + size_t size; + + size = talloc_get_size(vp->da); + + p = talloc_zero_array(vp, char, size); + da = (DICT_ATTR *) p; + talloc_set_type(p, DICT_ATTR); + memcpy(da, vp->da, size); + vp->da = da; + } +} + +/** Move pairs from source list to destination list respecting operator + * + * @note This function does some additional magic that's probably not needed + * in most places. Consider using radius_pairmove in server code. + * + * @note fr_pair_list_free should be called on the head of the source list to free + * unmoved attributes (if they're no longer needed). + * + * @note Does not respect tags when matching. + * + * @param[in] ctx for talloc + * @param[in,out] to destination list. + * @param[in,out] from source list. + * @param[in] op operator for move. + * + * @see radius_pairmove + */ +void fr_pair_list_move(TALLOC_CTX *ctx, VALUE_PAIR **to, VALUE_PAIR **from, FR_TOKEN op) +{ + VALUE_PAIR *i, *found; + VALUE_PAIR *head_new, **tail_new; + VALUE_PAIR *head_prepend; + VALUE_PAIR **tail_from; + + if (!to || !from || !*from) return; + + /* + * We're editing the "to" list while we're adding new + * attributes to it. We don't want the new attributes to + * be edited, so we create an intermediate list to hold + * them during the editing process. + */ + head_new = NULL; + tail_new = &head_new; + + /* + * Any attributes that are requested to be prepended + * are added to a temporary list here + */ + head_prepend = NULL; + + /* + * We're looping over the "from" list, moving some + * attributes out, but leaving others in place. + */ + tail_from = from; + while ((i = *tail_from) != NULL) { + VALUE_PAIR *j; + + VERIFY_VP(i); + + /* + * We never move Fall-Through. + */ + if (!i->da->vendor && i->da->attr == PW_FALL_THROUGH) { + tail_from = &(i->next); + continue; + } + + /* + * Unlike previous versions, we treat all other + * attributes as normal. i.e. there's no special + * treatment for passwords or Hint. + */ + + switch (i->op) { + /* + * Anything else are operators which + * shouldn't occur. We ignore them, and + * leave them in place. + */ + default: + tail_from = &(i->next); + continue; + + /* + * Add it to the "to" list, but only if + * it doesn't already exist. + */ + case T_OP_EQ: + found = fr_pair_find_by_da(*to, i->da, TAG_ANY); + if (!found) goto do_add; + + tail_from = &(i->next); + continue; + + /* + * Add it to the "to" list, and delete any attribute + * of the same vendor/attr which already exists. + */ + case T_OP_SET: + found = fr_pair_find_by_da(*to, i->da, TAG_ANY); + if (!found) goto do_add; + + /* + * Do NOT call fr_pair_delete_by_num() here, + * due to issues with re-writing + * "request->username". + * + * Everybody calls fr_pair_move, and + * expects it to work. We can't + * update request->username here, + * so instead we over-write the + * vp that it's pointing to. + */ + switch (found->da->type) { + default: + j = found->next; + memcpy(found, i, sizeof(*found)); + found->next = j; + break; + + case PW_TYPE_OCTETS: + fr_pair_value_memsteal(found, i->vp_octets); + i->vp_octets = NULL; + break; + + case PW_TYPE_STRING: + fr_pair_value_strsteal(found, i->vp_strvalue); + i->vp_strvalue = NULL; + found->tag = i->tag; + break; + } + + /* + * Delete *all* of the attributes + * of the same number. + */ + fr_pair_delete_by_num(&found->next, + found->da->attr, + found->da->vendor, TAG_ANY); + + /* + * Remove this attribute from the + * "from" list. + */ + *tail_from = i->next; + i->next = NULL; + fr_pair_list_free(&i); + continue; + + /* + * Move it from the old list and add it + * to the new list. + */ + case T_OP_ADD: + do_add: + *tail_from = i->next; + i->next = NULL; + *tail_new = i; + fr_pair_steal(ctx, i); + tail_new = &(i->next); + continue; + case T_OP_PREPEND: + i->next = head_prepend; + head_prepend = i; + fr_pair_steal(ctx, i); + continue; + } + } /* loop over the "from" list. */ + + /* + * If the op parameter was prepend, add the "new" list + * attributes first as those whose individual operator + * is prepend should be prepended to the resulting list + */ + if (op == T_OP_PREPEND) { + fr_pair_prepend(to, head_new); + } + + /* + * If there are any items in the prepend list prepend + * it to the "to" list + */ + fr_pair_prepend(to, head_prepend); + + /* + * If the op parameter was not prepend, take the "new" + * list, and append it to the "to" list. + */ + if (op != T_OP_PREPEND) { + fr_pair_add(to, head_new); + } +} + +/** Move matching pairs between VALUE_PAIR lists + * + * Move pairs of a matching attribute number, vendor number and tag from the + * the input list to the output list. + * + * @note fr_pair_list_free should be called on the head of the old list to free unmoved + attributes (if they're no longer needed). + * + * @param[in] ctx for talloc + * @param[in,out] to destination list. + * @param[in,out] from source list. + * @param[in] attr to match. If attribute PW_VENDOR_SPECIFIC and vendor 0, + * will match (and therefore copy) only VSAs. + * If attribute 0 and vendor 0 will match (and therefore copy) all + * attributes. + * @param[in] vendor to match. + * @param[in] tag to match, TAG_ANY matches any tag, TAG_NONE matches tagless VPs. + * @param[in] move if set to "true", VPs are moved. If set to "false", VPs are copied, and the old one deleted. + */ +static void fr_pair_list_move_by_num_internal(TALLOC_CTX *ctx, VALUE_PAIR **to, VALUE_PAIR **from, + unsigned int attr, unsigned int vendor, int8_t tag, + bool move) +{ + VALUE_PAIR *to_tail, *i, *next, *this; + VALUE_PAIR *iprev = NULL; + + /* + * Find the last pair in the "to" list and put it in "to_tail". + * + * @todo: replace the "if" with "VALUE_PAIR **tail" + */ + if (*to != NULL) { + to_tail = *to; + for(i = *to; i; i = i->next) { + VERIFY_VP(i); + to_tail = i; + } + } else + to_tail = NULL; + + /* + * Attr/vendor of 0 means "move them all". + * It's better than "fr_pair_add(foo,bar);bar=NULL" + */ + if ((vendor == 0) && (attr == 0)) { + if (*to) { + to_tail->next = *from; + } else { + *to = *from; + } + + for (i = *from; i; i = i->next) { + fr_pair_steal(ctx, i); + } + + *from = NULL; + return; + } + + for(i = *from; i; i = next) { + VERIFY_VP(i); + next = i->next; + + if (i->da->flags.has_tag && !TAG_EQ(tag, i->tag)) { + iprev = i; + continue; + } + + /* + * vendor=0, attr = PW_VENDOR_SPECIFIC means + * "match any vendor attribute". + */ + if ((vendor == 0) && (attr == PW_VENDOR_SPECIFIC)) { + /* + * It's a VSA: move it over. + */ + if (i->da->vendor != 0) goto move; + + /* + * It's Vendor-Specific: move it over. + */ + if (i->da->attr == attr) goto move; + + /* + * It's not a VSA: ignore it. + */ + iprev = i; + continue; + } + + /* + * If it isn't an exact match, ignore it. + */ + if (!((i->da->vendor == vendor) && (i->da->attr == attr))) { + iprev = i; + continue; + } + + move: + /* + * Remove the attribute from the "from" list. + */ + if (iprev) + iprev->next = next; + else + *from = next; + + if (move) { + this = i; + } else { + this = fr_pair_copy(ctx, i); + } + + /* + * Add the attribute to the "to" list. + */ + if (to_tail) + to_tail->next = this; + else + *to = this; + to_tail = this; + this->next = NULL; + + if (move) { + fr_pair_steal(ctx, i); + } else { + talloc_free(i); + } + } +} + + +/** Move matching pairs between VALUE_PAIR lists + * + * Move pairs of a matching attribute number, vendor number and tag from the + * the input list to the output list. + * + * @note pairs which are moved have their parent changed to ctx. + * + * @note fr_pair_list_free should be called on the head of the old list to free unmoved + attributes (if they're no longer needed). + * + * @param[in] ctx for talloc + * @param[in,out] to destination list. + * @param[in,out] from source list. + * @param[in] attr to match. If attribute PW_VENDOR_SPECIFIC and vendor 0, + * will match (and therefore copy) only VSAs. + * If attribute 0 and vendor 0 will match (and therefore copy) all + * attributes. + * @param[in] vendor to match. + * @param[in] tag to match, TAG_ANY matches any tag, TAG_NONE matches tagless VPs. + */ +void fr_pair_list_move_by_num(TALLOC_CTX *ctx, VALUE_PAIR **to, VALUE_PAIR **from, + unsigned int attr, unsigned int vendor, int8_t tag) +{ + fr_pair_list_move_by_num_internal(ctx, to, from, attr, vendor, tag, true); +} + + +/** Copy / delete matching pairs between VALUE_PAIR lists + * + * Move pairs of a matching attribute number, vendor number and tag from the + * the input list to the output list. Like fr_pair_list_move_by_num(), but + * instead does copy / delete. + * + * @note The pair is NOT reparented. It is copied and deleted. + * + * @note fr_pair_list_free should be called on the head of the old list to free unmoved + attributes (if they're no longer needed). + * + * @param[in] ctx for talloc + * @param[in,out] to destination list. + * @param[in,out] from source list. + * @param[in] attr to match. If attribute PW_VENDOR_SPECIFIC and vendor 0, + * will match (and therefore copy) only VSAs. + * If attribute 0 and vendor 0 will match (and therefore copy) all + * attributes. + * @param[in] vendor to match. + * @param[in] tag to match, TAG_ANY matches any tag, TAG_NONE matches tagless VPs. + */ +void fr_pair_list_mcopy_by_num(TALLOC_CTX *ctx, VALUE_PAIR **to, VALUE_PAIR **from, + unsigned int attr, unsigned int vendor, int8_t tag) +{ + fr_pair_list_move_by_num_internal(ctx, to, from, attr, vendor, tag, false); +} + + +/** Convert string value to native attribute value + * + * @param vp to assign value to. + * @param value string to convert. Binary safe for variable length values if len is provided. + * @param inlen may be < 0 in which case strlen(len) is used to determine length, else inline + * should be the length of the string or sub string to parse. + * @return 0 on success -1 on error. + */ +int fr_pair_value_from_str(VALUE_PAIR *vp, char const *value, size_t inlen) +{ + ssize_t ret; + PW_TYPE type; + VERIFY_VP(vp); + + if (!value) return -1; + + type = vp->da->type; + + /* + * We presume that the input data is from a double quoted + * string, and needs escaping + */ + ret = value_data_from_str(vp, &vp->data, &type, vp->da, value, inlen, '"'); + if (ret < 0) return -1; + + /* + * If we parsed to a different type than the DA associated with + * the VALUE_PAIR we now need to fixup the DA. + */ + if (type != vp->da->type) { + DICT_ATTR const *da; + + da = dict_attrbytype(vp->da->attr, vp->da->vendor, type); + if (!da) { + fr_strerror_printf("Cannot find %s variant of attribute \"%s\"", + fr_int2str(dict_attr_types, type, "<INVALID>"), vp->da->name); + return -1; + } + vp->da = da; + } + + vp->vp_length = ret; + vp->type = VT_DATA; + + VERIFY_VP(vp); + + return 0; +} + +/** Use simple heuristics to create an VALUE_PAIR from an unknown address string + * + * If a DICT_ATTR is not provided for the address type, parsing will fail with + * and error. + * + * @param ctx to allocate VP in. + * @param value IPv4/IPv6 address/prefix string. + * @param ipv4 dictionary attribute to use for an IPv4 address. + * @param ipv6 dictionary attribute to use for an IPv6 address. + * @param ipv4_prefix dictionary attribute to use for an IPv4 prefix. + * @param ipv6_prefix dictionary attribute to use for an IPv6 prefix. + * @return NULL on error, or new VALUE_PAIR. + */ +VALUE_PAIR *fr_pair_afrom_ip_str(TALLOC_CTX *ctx, char const *value, DICT_ATTR *ipv4, DICT_ATTR *ipv6, + DICT_ATTR *ipv4_prefix, DICT_ATTR *ipv6_prefix) +{ + VALUE_PAIR *vp; + DICT_ATTR *da = NULL; + + if (!fr_assert(ipv4 || ipv6 || ipv4_prefix || ipv6_prefix)) { + return NULL; + } + + /* No point in repeating the work of fr_pair_value_from_str */ + if (strchr(value, ':')) { + if (strchr(value, '/')) { + da = ipv6_prefix; + goto finish; + } + + da = ipv6; + goto finish; + } + + if (strchr(value, '/')) { + da = ipv4_prefix; + goto finish; + } + + if (ipv4) { + da = ipv4; + goto finish; + } + + fr_strerror_printf("Invalid IP value specified, allowed types are %s%s%s%s", + ipv4 ? "ipaddr " : "", ipv6 ? "ipv6addr " : "", + ipv4_prefix ? "ipv4prefix " : "", ipv6_prefix ? "ipv6prefix" : ""); + +finish: + vp = fr_pair_afrom_da(ctx, da); + if (!vp) return NULL; + if (fr_pair_value_from_str(vp, value, -1) < 0) { + talloc_free(vp); + return NULL; + } + + return vp; +} + + +/** Create a valuepair from an ASCII attribute and value + * + * Where the attribute name is in the form: + * - Attr-%d + * - Attr-%d.%d.%d... + * - Vendor-%d-Attr-%d + * - VendorName-Attr-%d + * + * @param ctx for talloc + * @param attribute name to parse. + * @param value to parse (must be a hex string). + * @param op to assign to new valuepair. + * @return new valuepair or NULL on error. + */ +static VALUE_PAIR *fr_pair_make_unknown(TALLOC_CTX *ctx, + char const *attribute, char const *value, + FR_TOKEN op) +{ + VALUE_PAIR *vp, *vp2; + DICT_ATTR const *da; + + uint8_t *data; + size_t size; + ssize_t len; + + vp = fr_pair_alloc(ctx); + if (!vp) return NULL; + + vp->da = dict_unknown_afrom_str(vp, attribute); + if (!vp->da) { + talloc_free(vp); + return NULL; + } + + /* + * No value. Nothing more to do. + */ + if (!value) return vp; + + /* + * Unknown attributes MUST be of type 'octets' + */ + if (strncasecmp(value, "0x", 2) != 0) { + fr_strerror_printf("Unknown attribute \"%s\" requires a hex " + "string, not \"%s\"", attribute, value); + talloc_free(vp); + return NULL; + } + + /* + * Convert the hex data to binary. + */ + size = strlen(value + 2); + + vp->vp_length = size >> 1; + vp->vp_octets = data = talloc_array(vp, uint8_t, vp->vp_length); + vp->type = VT_DATA; + vp->op = (op == 0) ? T_OP_EQ : op; + + if (fr_hex2bin(data, vp->vp_length, value + 2, size) != vp->vp_length) { + fr_strerror_printf("Invalid hex string"); + talloc_free(vp); + return NULL; + } + + /* + * It's still unknown, return it as-is. + */ + da = dict_attrbyvalue(vp->da->attr, vp->da->vendor); + if (!da) return vp; + + /* + * It MIGHT be known. See if we can decode the raw data + * into a valid attribute. + */ + len = data2vp(ctx, NULL, NULL, NULL, da, + vp->vp_octets, vp->vp_length, vp->vp_length, + &vp2); + if (len <= 0) return vp; + + /* + * It's still unknown. Return the original VP. + */ + if (vp2->da->flags.is_unknown) { + fr_pair_list_free(&vp2); + return vp; + } + + /* + * Didn't parse all of it. Return the "unknown" one. + * + * FIXME: it COULD have parsed 2 attributes and + * then not the third, so returning 2 "knowns" + * and 1 "unknown" is likely preferable. + */ + if ((size_t) len < vp->vp_length) { + fr_pair_list_free(&vp2); + return vp; + } + + fr_pair_list_free(&vp); + return vp2; +} + + +/** Create a VALUE_PAIR from ASCII strings + * + * Converts an attribute string identifier (with an optional tag qualifier) + * and value string into a VALUE_PAIR. + * + * The string value is parsed according to the type of VALUE_PAIR being created. + * + * @param[in] ctx for talloc + * @param[in] vps list where the attribute will be added (optional) + * @param[in] attribute name. + * @param[in] value attribute value (may be NULL if value will be set later). + * @param[in] op to assign to new VALUE_PAIR. + * @return a new VALUE_PAIR. + */ +VALUE_PAIR *fr_pair_make(TALLOC_CTX *ctx, VALUE_PAIR **vps, + char const *attribute, char const *value, FR_TOKEN op) +{ + DICT_ATTR const *da; + VALUE_PAIR *vp; + char *tc, *ts; + int8_t tag; + bool found_tag; + char buffer[256]; + char const *attrname = attribute; + + /* + * Check for tags in 'Attribute:Tag' format. + */ + found_tag = false; + tag = TAG_ANY; + + ts = strrchr(attribute, ':'); + if (ts && !ts[1]) { + fr_strerror_printf("Invalid tag for attribute %s", attribute); + return NULL; + } + + if (ts && ts[1]) { + strlcpy(buffer, attribute, sizeof(buffer)); + attrname = buffer; + ts = strrchr(attrname, ':'); + if (!ts) return NULL; + + /* Colon found with something behind it */ + if (ts[1] == '*' && ts[2] == 0) { + /* Wildcard tag for check items */ + tag = TAG_ANY; + *ts = '\0'; + } else if ((ts[1] >= '0') && (ts[1] <= '9')) { + /* It's not a wild card tag */ + tag = strtol(ts + 1, &tc, 0); + if (tc && !*tc && TAG_VALID_ZERO(tag)) + *ts = '\0'; + else tag = TAG_ANY; + } else { + fr_strerror_printf("Invalid tag for attribute %s", attribute); + return NULL; + } + found_tag = true; + } + + /* + * It's not found in the dictionary, so we use + * another method to create the attribute. + */ + da = dict_attrbyname(attrname); + if (!da) { + vp = fr_pair_make_unknown(ctx, attrname, value, op); + if (vp && vps) fr_pair_add(vps, vp); + return vp; + } + + /* Check for a tag in the 'Merit' format of: + * :Tag:Value. Print an error if we already found + * a tag in the Attribute. + */ + + if (value && (*value == ':' && da->flags.has_tag)) { + /* If we already found a tag, this is invalid */ + if(found_tag) { + fr_strerror_printf("Duplicate tag %s for attribute %s", + value, da->name); + DEBUG("Duplicate tag %s for attribute %s\n", + value, da->name); + return NULL; + } + /* Colon found and attribute allows a tag */ + if (value[1] == '*' && value[2] == ':') { + /* Wildcard tag for check items */ + tag = TAG_ANY; + value += 3; + } else { + /* Real tag */ + tag = strtol(value + 1, &tc, 0); + if (tc && *tc==':' && TAG_VALID_ZERO(tag)) + value = tc + 1; + else tag = 0; + } + } + + vp = fr_pair_afrom_da(ctx, da); + if (!vp) return NULL; + vp->op = (op == 0) ? T_OP_EQ : op; + vp->tag = tag; + + switch (vp->op) { + case T_OP_CMP_TRUE: + case T_OP_CMP_FALSE: + vp->vp_strvalue = NULL; + vp->vp_length = 0; + value = NULL; /* ignore it! */ + break; + + /* + * Regular expression comparison of integer attributes + * does a STRING comparison of the names of their + * integer attributes. + */ + case T_OP_REG_EQ: /* =~ */ + case T_OP_REG_NE: /* !~ */ + { +#ifndef HAVE_REGEX + fr_strerror_printf("Regular expressions are not supported"); + return NULL; +#else + ssize_t slen; + regex_t *preg; + + /* + * Someone else will fill in the value. + */ + if (!value) break; + + talloc_free(vp); + + slen = regex_compile(ctx, &preg, value, strlen(value), false, false, false, true); + if (slen <= 0) { + fr_strerror_printf("Error at offset %zu compiling regex for %s: %s", -slen, + attribute, fr_strerror()); + return NULL; + } + talloc_free(preg); + + vp = fr_pair_make(ctx, NULL, attribute, NULL, op); + if (!vp) return NULL; + + if (fr_pair_mark_xlat(vp, value) < 0) { + talloc_free(vp); + return NULL; + } + + value = NULL; /* ignore it */ + break; +#endif + } + default: + break; + } + + /* + * We allow this for stupidity, but it's really a bad idea. + */ + if (vp->da->type == PW_TYPE_TLV) { + ssize_t len; + DICT_ATTR const *unknown; + VALUE_PAIR *head = NULL; + VALUE_PAIR **tail = &head; + + if (!value) { + talloc_free(vp); + return NULL; + } + + unknown = dict_unknown_afrom_fields(vp, vp->da->attr, vp->da->vendor); + if (!unknown) { + talloc_free(vp); + return NULL; + } + + vp->da = unknown; + + /* + * Parse it as an unknown type, i.e. octets. + */ + if (fr_pair_value_from_str(vp, value, -1) < 0) { + talloc_free(vp); + return NULL; + } + + /* + * It's badly formatted. Treat it as unknown. + */ + if (rad_tlv_ok(vp->vp_octets, vp->vp_length, 1, 1) < 0) { + goto do_add; + } + + /* + * Decode the TLVs + */ + len = rad_data2vp_tlvs(ctx, NULL, NULL, NULL, da, vp->vp_octets, + vp->vp_length, tail); + if (len < 0) { + goto do_add; + } + + talloc_free(vp); + vp = head; + goto do_add; + } + + /* + * FIXME: if (strcasecmp(attribute, vp->da->name) != 0) + * then the user MAY have typed in the attribute name + * as Vendor-%d-Attr-%d, and the value MAY be octets. + * + * We probably want to fix fr_pair_value_from_str to accept + * octets as values for any attribute. + */ + if (value && (fr_pair_value_from_str(vp, value, -1) < 0)) { + talloc_free(vp); + return NULL; + } + +do_add: + if (vps) fr_pair_add(vps, vp); + return vp; +} + +/** Mark a valuepair for xlat expansion + * + * Copies xlat source (unprocessed) string to valuepair value, + * and sets value type. + * + * @param vp to mark for expansion. + * @param value to expand. + * @return 0 if marking succeeded or -1 if vp already had a value, or OOM. + */ +int fr_pair_mark_xlat(VALUE_PAIR *vp, char const *value) +{ + char *raw; + + /* + * valuepair should not already have a value. + */ + if (vp->type != VT_NONE) { + fr_strerror_printf("Pair already has a value"); + return -1; + } + + raw = talloc_typed_strdup(vp, value); + if (!raw) { + fr_strerror_printf("Out of memory"); + return -1; + } + + vp->type = VT_XLAT; + vp->value.xlat = raw; + vp->vp_length = 0; + + return 0; +} + + +/** Read a single valuepair from a buffer, and advance the pointer + * + * Returns T_EOL if end of line was encountered. + * + * @param[in,out] ptr to read from and update. + * @param[out] raw The struct to write the raw VALUE_PAIR to. + * @return the last token read. + */ +FR_TOKEN fr_pair_raw_from_str(char const **ptr, VALUE_PAIR_RAW *raw) +{ + char const *p; + char *q; + FR_TOKEN ret = T_INVALID, next, quote; + char buf[8]; + + if (!ptr || !*ptr || !raw) { + fr_strerror_printf("Invalid arguments"); + return T_INVALID; + } + + /* + * Skip leading spaces + */ + p = *ptr; + while ((*p == ' ') || (*p == '\t')) p++; + + if (!*p) { + fr_strerror_printf("No token read where we expected " + "an attribute name"); + return T_INVALID; + } + + if (*p == '#') return T_HASH; + + /* + * Try to get the attribute name. + */ + q = raw->l_opand; + *q = '\0'; + while (*p) { + uint8_t const *t = (uint8_t const *) p; + + if (q >= (raw->l_opand + sizeof(raw->l_opand))) { + too_long: + fr_strerror_printf("Attribute name too long"); + return T_INVALID; + } + + /* + * This is arguably easier than trying to figure + * out which operators come after the attribute + * name. Yes, our "lexer" is bad. + */ + if (!dict_attr_allowed_chars[(unsigned int) *t]) { + break; + } + + /* + * Attribute:=value is NOT + * + * Attribute: + * = + * value + */ + if ((*p == ':') && (!isdigit((int) p[1]))) { + break; + } + + *(q++) = *(p++); + } + + /* + * Haven't found any valid characters in the name. + */ + if (!*raw->l_opand) { + fr_strerror_printf("Invalid attribute name"); + return T_INVALID; + } + + /* + * Look for tag (:#). This is different from :=, which + * is an operator. + */ + if ((*p == ':') && (isdigit((int) p[1]))) { + if (q >= (raw->l_opand + sizeof(raw->l_opand))) { + goto too_long; + } + *(q++) = *(p++); + + while (isdigit((int) *p)) { + if (q >= (raw->l_opand + sizeof(raw->l_opand))) { + goto too_long; + } + *(q++) = *(p++); + } + } + + *q = '\0'; + *ptr = p; + + /* Now we should have an operator here. */ + raw->op = gettoken(ptr, buf, sizeof(buf), false); + if (raw->op < T_EQSTART || raw->op > T_EQEND) { + fr_strerror_printf("Expecting operator"); + + return T_INVALID; + } + + /* + * Read value. Note that empty string values are allowed + */ + quote = gettoken(ptr, raw->r_opand, sizeof(raw->r_opand), false); + if (quote == T_EOL) { + fr_strerror_printf("Failed to get value"); + + return T_INVALID; + } + + /* + * Peek at the next token. Must be T_EOL, T_COMMA, or T_HASH + */ + p = *ptr; + + next = gettoken(&p, buf, sizeof(buf), false); + switch (next) { + case T_HASH: + next = T_EOL; + break; + + case T_EOL: + break; + + case T_COMMA: + *ptr = p; + break; + + default: + fr_strerror_printf("Expected end of line or comma"); + return T_INVALID; + } + ret = next; + + switch (quote) { + /* + * Perhaps do xlat's + */ + case T_DOUBLE_QUOTED_STRING: + /* + * Only report as double quoted if it contained valid + * a valid xlat expansion. + */ + p = strchr(raw->r_opand, '%'); + if (p && (p[1] == '{')) { + raw->quote = quote; + } else { + raw->quote = T_SINGLE_QUOTED_STRING; + } + + break; + + case T_SINGLE_QUOTED_STRING: + case T_BACK_QUOTED_STRING: + case T_BARE_WORD: + raw->quote = quote; + break; + + default: + fr_strerror_printf("Failed to find expected value on right hand side in %s", raw->l_opand); + return T_INVALID; + } + + return ret; +} + +/** Read one line of attribute/value pairs into a list. + * + * The line may specify multiple attributes separated by commas. + * + * @note If the function returns T_INVALID, an error has occurred and + * @note the valuepair list should probably be freed. + * + * @param ctx for talloc + * @param buffer to read valuepairs from. + * @param list where the parsed VALUE_PAIRs will be appended. + * @return the last token parsed, or T_INVALID + */ +FR_TOKEN fr_pair_list_afrom_str(TALLOC_CTX *ctx, char const *buffer, VALUE_PAIR **list) +{ + VALUE_PAIR *vp, *head, **tail; + char const *p; + FR_TOKEN last_token = T_INVALID; + VALUE_PAIR_RAW raw; + + /* + * We allow an empty line. + */ + if (buffer[0] == 0) { + return T_EOL; + } + + head = NULL; + tail = &head; + + p = buffer; + do { + raw.l_opand[0] = '\0'; + raw.r_opand[0] = '\0'; + + last_token = fr_pair_raw_from_str(&p, &raw); + + /* + * JUST a hash. Don't try to create a VP. + * Let the caller determine if an empty list is OK. + */ + if (last_token == T_HASH) { + last_token = T_EOL; + break; + } + if (last_token == T_INVALID) break; + + if (raw.quote == T_DOUBLE_QUOTED_STRING) { + vp = fr_pair_make(ctx, NULL, raw.l_opand, NULL, raw.op); + if (!vp) { + last_token = T_INVALID; + break; + } + if (fr_pair_mark_xlat(vp, raw.r_opand) < 0) { + talloc_free(vp); + last_token = T_INVALID; + break; + } + } else { + vp = fr_pair_make(ctx, NULL, raw.l_opand, raw.r_opand, raw.op); + if (!vp) { + last_token = T_INVALID; + break; + } + } + + *tail = vp; + tail = &((*tail)->next); + } while (*p && (last_token == T_COMMA)); + + if (last_token == T_INVALID) { + fr_pair_list_free(&head); + } else { + fr_pair_add(list, head); + } + + /* + * And return the last token which we read. + */ + return last_token; +} + +/* + * Read valuepairs from the fp up to End-Of-File. + */ +int fr_pair_list_afrom_file(TALLOC_CTX *ctx, VALUE_PAIR **out, FILE *fp, bool *pfiledone) +{ + char buf[8192]; + FR_TOKEN last_token = T_EOL; + + vp_cursor_t cursor; + + VALUE_PAIR *vp = NULL; + fr_cursor_init(&cursor, out); + + while (fgets(buf, sizeof(buf), fp) != NULL) { + /* + * If we get a '\n' by itself, we assume that's + * the end of that VP + */ + if (buf[0] == '\n') { + if (vp) { + *pfiledone = false; + return 0; + } + continue; + } + + /* + * Comments get ignored + */ + if (buf[0] == '#') continue; + + /* + * Read all of the attributes on the current line. + */ + vp = NULL; + last_token = fr_pair_list_afrom_str(ctx, buf, &vp); + if (!vp) { + if (last_token != T_EOL) goto error; + break; + } + + fr_cursor_merge(&cursor, vp); + buf[0] = '\0'; + } + *pfiledone = true; + + return 0; + +error: + *pfiledone = false; + vp = fr_cursor_first(&cursor); + if (vp) fr_pair_list_free(&vp); + + return -1; +} + +/** Compare two pairs, using the operator from "a" + * + * i.e. given two attributes, it does: + * + * (b->data) (a->operator) (a->data) + * + * e.g. "foo" != "bar" + * + * @param[in] a the first attribute + * @param[in] b the second attribute + * @return 1 if true, 0 if false, -1 on error. + */ +int fr_pair_cmp(VALUE_PAIR *a, VALUE_PAIR *b) +{ + if (!a) return -1; + + VERIFY_VP(a); + if (b) VERIFY_VP(b); + + switch (a->op) { + case T_OP_CMP_TRUE: + return (b != NULL); + + case T_OP_CMP_FALSE: + return (b == NULL); + + /* + * a is a regex, compile it, print b to a string, + * and then do string comparisons. + */ + case T_OP_REG_EQ: + case T_OP_REG_NE: +#ifndef HAVE_REGEX + return -1; +#else + if (!b) return false; + + { + ssize_t slen; + regex_t *preg; + char *value; + + if (!fr_assert(a->da->type == PW_TYPE_STRING)) return -1; + + slen = regex_compile(NULL, &preg, a->value.xlat, talloc_array_length(a->value.xlat) - 1, false, false, false, true); + if (slen <= 0) { + fr_strerror_printf("Error at offset %zu compiling regex for %s: %s", + -slen, a->da->name, fr_strerror()); + return -1; + } + value = vp_aprints_value(NULL, b, '\0'); + if (!value) { + talloc_free(preg); + return -1; + } + + /* + * Don't care about substring matches, oh well... + */ + slen = regex_exec(preg, value, talloc_array_length(value) - 1, NULL, NULL); + talloc_free(preg); + talloc_free(value); + + if (slen < 0) return -1; + if (a->op == T_OP_REG_EQ) return (int)slen; + return !slen; + } +#endif + + default: /* we're OK */ + if (!b) return false; + break; + } + + return fr_pair_cmp_op(a->op, b, a); +} + +/** Determine equality of two lists + * + * This is useful for comparing lists of attributes inserted into a binary tree. + * + * @param a first list of VALUE_PAIRs. + * @param b second list of VALUE_PAIRs. + * @return -1 if a < b, 0 if the two lists are equal, 1 if a > b, -2 on error. + */ +int fr_pair_list_cmp(VALUE_PAIR *a, VALUE_PAIR *b) +{ + vp_cursor_t a_cursor, b_cursor; + VALUE_PAIR *a_p, *b_p; + int ret; + + for (a_p = fr_cursor_init(&a_cursor, &a), b_p = fr_cursor_init(&b_cursor, &b); + a_p && b_p; + a_p = fr_cursor_next(&a_cursor), b_p = fr_cursor_next(&b_cursor)) { + /* Same VP, no point doing expensive checks */ + if (a_p == b_p) { + continue; + } + + if (a_p->da < b_p->da) { + return -1; + } + if (a_p->da > b_p->da) { + return 1; + } + + if (a_p->tag < b_p->tag) { + return -1; + } + if (a_p->tag > b_p->tag) { + return 1; + } + + ret = value_data_cmp(a_p->da->type, &a_p->data, a_p->vp_length, + b_p->da->type, &b_p->data, b_p->vp_length); + if (ret != 0) { + fr_assert(ret >= -1); /* Comparison error */ + return ret; + } + } + + if (!a_p && !b_p) { + return 0; + } + + if (!a_p) { + return -1; + } + + /* if(!b_p) */ + return 1; +} + +/** Set the type of the VALUE_PAIR value buffer to match it's DICT_ATTR + * + * @param vp to fixup. + */ +static void fr_pair_value_set_type(VALUE_PAIR *vp) +{ + if (!vp->data.ptr) return; + + switch (vp->da->type) { + case PW_TYPE_OCTETS: + talloc_set_type(vp->data.ptr, uint8_t); + return; + + case PW_TYPE_STRING: + talloc_set_type(vp->data.ptr, char); + return; + + default: + return; + } +} + +/** Copy data into an "octets" data type. + * + * @param[in,out] vp to update + * @param[in] src data to copy + * @param[in] size of the data, may be 0 in which case previous value will be freed. + */ +void fr_pair_value_memcpy(VALUE_PAIR *vp, uint8_t const *src, size_t size) +{ + uint8_t *p = NULL, *q; + + VERIFY_VP(vp); + + if (size > 0) { + p = talloc_memdup(vp, src, size); + if (!p) return; + talloc_set_type(p, uint8_t); + } + + memcpy(&q, &vp->vp_octets, sizeof(q)); + TALLOC_FREE(q); + + vp->vp_octets = p; + vp->vp_length = size; + + if (size > 0) fr_pair_value_set_type(vp); +} + +/** Reparent an allocated octet buffer to a VALUE_PAIR + * + * @param[in,out] vp to update + * @param[in] src buffer to steal. + */ +void fr_pair_value_memsteal(VALUE_PAIR *vp, uint8_t const *src) +{ + uint8_t *q; + + VERIFY_VP(vp); + + memcpy(&q, &vp->vp_octets, sizeof(q)); + talloc_free(q); + + vp->vp_octets = talloc_steal(vp, src); + vp->type = VT_DATA; + vp->vp_length = talloc_array_length(vp->vp_strvalue); + fr_pair_value_set_type(vp); +} + +/** Reparent an allocated char buffer to a VALUE_PAIR + * + * @param[in,out] vp to update + * @param[in] src buffer to steal. + */ +void fr_pair_value_strsteal(VALUE_PAIR *vp, char const *src) +{ + uint8_t *q; + + VERIFY_VP(vp); + + memcpy(&q, &vp->vp_octets, sizeof(q)); + talloc_free(q); + + vp->vp_strvalue = talloc_steal(vp, src); + vp->type = VT_DATA; + vp->vp_length = talloc_array_length(vp->vp_strvalue) - 1; + fr_pair_value_set_type(vp); +} + +/** Copy data into an "string" data type. + * + * @param[in,out] vp to update + * @param[in] src data to copy + */ +void fr_pair_value_strcpy(VALUE_PAIR *vp, char const *src) +{ + char *p, *q; + + VERIFY_VP(vp); + + p = talloc_strdup(vp, src); + + if (!p) return; + + memcpy(&q, &vp->vp_strvalue, sizeof(q)); + talloc_free(q); + + vp->vp_strvalue = p; + vp->type = VT_DATA; + vp->vp_length = talloc_array_length(vp->vp_strvalue) - 1; + fr_pair_value_set_type(vp); +} + +/** Copy data into an "string" data type. + * + * @note unlike the original strncpy, this function does not stop + * if it finds \0 bytes embedded in the string. + * + * @param[in,out] vp to update. + * @param[in] src data to copy. + * @param[in] len of data to copy. + */ +void fr_pair_value_bstrncpy(VALUE_PAIR *vp, void const *src, size_t len) +{ + char *p, *q; + + VERIFY_VP(vp); + + p = talloc_array(vp, char, len + 1); + if (!p) return; + + memcpy(p, src, len); /* embdedded \0 safe */ + p[len] = '\0'; + + memcpy(&q, &vp->vp_strvalue, sizeof(q)); + talloc_free(q); + + vp->vp_strvalue = p; + vp->type = VT_DATA; + vp->vp_length = len; + fr_pair_value_set_type(vp); +} + +/** Print data into an "string" data type. + * + * @param[in,out] vp to update + * @param[in] fmt the format string + */ +void fr_pair_value_sprintf(VALUE_PAIR *vp, char const *fmt, ...) +{ + va_list ap; + char *p, *q; + + VERIFY_VP(vp); + + va_start(ap, fmt); + p = talloc_vasprintf(vp, fmt, ap); + va_end(ap); + + if (!p) return; + + memcpy(&q, &vp->vp_strvalue, sizeof(q)); + talloc_free(q); + + vp->vp_strvalue = p; + vp->type = VT_DATA; + + vp->vp_length = talloc_array_length(vp->vp_strvalue) - 1; + fr_pair_value_set_type(vp); +} + +#ifdef WITH_VERIFY_PTR +/* + * Verify a VALUE_PAIR + */ +inline void fr_pair_verify(char const *file, int line, VALUE_PAIR const *vp) +{ + if (!vp) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR pointer was NULL", file, line); + fr_assert(0); + fr_exit_now(1); + } + + (void) talloc_get_type_abort(vp, VALUE_PAIR); + + if (!vp->da) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR da pointer was NULL", file, line); + fr_assert(0); + fr_exit_now(1); + } + + if (vp->data.ptr) switch (vp->da->type) { + case PW_TYPE_OCTETS: + { + size_t len; + TALLOC_CTX *parent; + + if (!talloc_get_type(vp->data.ptr, uint8_t)) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR \"%s\" data buffer type should be " + "uint8_t but is %s\n", file, line, vp->da->name, talloc_get_name(vp->data.ptr)); + (void) talloc_get_type_abort(vp->data.ptr, uint8_t); + } + + len = talloc_array_length(vp->vp_octets); + if (vp->vp_length > len) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR \"%s\" length %zu is greater than " + "uint8_t data buffer length %zu\n", file, line, vp->da->name, vp->vp_length, len); + fr_assert(0); + fr_exit_now(1); + } + + parent = talloc_parent(vp->data.ptr); + if (parent != vp) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR \"%s\" char buffer is not " + "parented by VALUE_PAIR %p, instead parented by %p (%s)\n", + file, line, vp->da->name, + vp, parent, parent ? talloc_get_name(parent) : "NULL"); + fr_assert(0); + fr_exit_now(1); + } + } + break; + + case PW_TYPE_STRING: + { + size_t len; + TALLOC_CTX *parent; + + if (!talloc_get_type(vp->data.ptr, char)) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR \"%s\" data buffer type should be " + "char but is %s\n", file, line, vp->da->name, talloc_get_name(vp->data.ptr)); + (void) talloc_get_type_abort(vp->data.ptr, char); + } + + len = (talloc_array_length(vp->vp_strvalue) - 1); + if (vp->vp_length > len) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR \"%s\" length %zu is greater than " + "char buffer length %zu\n", file, line, vp->da->name, vp->vp_length, len); + fr_assert(0); + fr_exit_now(1); + } + + if (vp->vp_strvalue[vp->vp_length] != '\0') { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR \"%s\" char buffer not \\0 " + "terminated\n", file, line, vp->da->name); + fr_assert(0); + fr_exit_now(1); + } + + parent = talloc_parent(vp->data.ptr); + if (parent != vp) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR \"%s\" char buffer is not " + "parented by VALUE_PAIR %p, instead parented by %p (%s)\n", + file, line, vp->da->name, + vp, parent, parent ? talloc_get_name(parent) : "NULL"); + fr_assert(0); + fr_exit_now(1); + } + } + break; + + default: + break; + } + + if (vp->da->flags.is_unknown) { + (void) talloc_get_type_abort(vp->da, DICT_ATTR); + } else { + DICT_ATTR const *da; + + /* + * Attribute may be present with multiple names + */ + da = dict_attrbyname(vp->da->name); + if (!da) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR attribute %p \"%s\" (%s) " + "not found in global dictionary", + file, line, vp->da, vp->da->name, + fr_int2str(dict_attr_types, vp->da->type, "<INVALID>")); + fr_assert(0); + fr_exit_now(1); + } + + if (da->type == PW_TYPE_COMBO_IP_ADDR) { + da = dict_attrbytype(vp->da->attr, vp->da->vendor, vp->da->type); + if (!da) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR attribute %p \"%s\" " + "variant (%s) not found in global dictionary", + file, line, vp->da, vp->da->name, + fr_int2str(dict_attr_types, vp->da->type, "<INVALID>")); + fr_assert(0); + fr_exit_now(1); + } + } + + + if (da != vp->da) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: VALUE_PAIR " + "dictionary pointer %p \"%s\" (%s) " + "and global dictionary pointer %p \"%s\" (%s) differ", + file, line, vp->da, vp->da->name, + fr_int2str(dict_attr_types, vp->da->type, "<INVALID>"), + da, da->name, fr_int2str(dict_attr_types, da->type, "<INVALID>")); + fr_assert(0); + fr_exit_now(1); + } + } +} + +/* + * Verify a pair list + */ +void fr_pair_list_verify(char const *file, int line, TALLOC_CTX *expected, VALUE_PAIR *vps, char const *name) +{ + vp_cursor_t cursor; + VALUE_PAIR *vp; + TALLOC_CTX *parent; + + for (vp = fr_cursor_init(&cursor, &vps); + vp; + vp = fr_cursor_next(&cursor)) { + VERIFY_VP(vp); + + parent = talloc_parent(vp); + if (expected && (parent != expected)) { + FR_FAULT_LOG("CONSISTENCY CHECK FAILED %s[%u]: Expected VALUE_PAIR \"%s\" to be parented " + "by %p (%s) name %s, instead parented by %p (%s)\n", + file, line, vp->da->name, + expected, talloc_get_name(expected), name, + parent, parent ? talloc_get_name(parent) : "NULL"); + + fr_log_talloc_report(expected); + if (parent) fr_log_talloc_report(parent); + + fr_assert(0); + fr_exit_now(1); + } + + } +} +#endif |