Skip to content

MONGOCRYPT-762 Generate text search token sets from StrEncode output #946

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 8 commits into from
Feb 3, 2025
Merged

MONGOCRYPT-762 Generate text search token sets from StrEncode output #946

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
da19785
MONGOCRYPT-762 Generate text search token sets from StrEncode output
erwee Jan 28, 2025
1450969
assert safe cast of uint32_t length to int
erwee Jan 29, 2025
22a1e32
Fix memleak
erwee Jan 29, 2025
867d146
Fix conversion errors
erwee Jan 29, 2025
dacfaa1
Gabe's suggestions
erwee Jan 29, 2025
348ab30
Gabe's feedback 2; Fix empty sets bug and add test
erwee Jan 29, 2025
3f6df0d
revert back to double braces
erwee Jan 31, 2025
1d95bf4
fixes
erwee Jan 31, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion src/mc-fle2-insert-update-payload-private-v2.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,8 @@
_mongocrypt_buffer_t encryptedTokens; \
} mc_Text##Type##TokenSet_t; \
void mc_Text##Type##TokenSet_init(mc_Text##Type##TokenSet_t *); \
void mc_Text##Type##TokenSet_cleanup(mc_Text##Type##TokenSet_t *)
void mc_Text##Type##TokenSet_cleanup(mc_Text##Type##TokenSet_t *); \
void mc_Text##Type##TokenSet_copy(const mc_Text##Type##TokenSet_t *src, mc_Text##Type##TokenSet_t *dest)

DEF_TEXT_SEARCH_TOKEN_SET(Exact);
DEF_TEXT_SEARCH_TOKEN_SET(Substring);
Expand Down
8 changes: 8 additions & 0 deletions src/mc-fle2-insert-update-payload-v2.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,14 @@
_mongocrypt_buffer_cleanup(&ts->escDerivedToken); \
_mongocrypt_buffer_cleanup(&ts->serverDerivedFromDataToken); \
_mongocrypt_buffer_cleanup(&ts->encryptedTokens); \
} \
void mc_Text##Type##TokenSet_copy(const mc_Text##Type##TokenSet_t *src, mc_Text##Type##TokenSet_t *dst) { \
BSON_ASSERT_PARAM(src); \
BSON_ASSERT_PARAM(dst); \
_mongocrypt_buffer_copy_to(&src->edcDerivedToken, &dst->edcDerivedToken); \
_mongocrypt_buffer_copy_to(&src->escDerivedToken, &dst->escDerivedToken); \
_mongocrypt_buffer_copy_to(&src->serverDerivedFromDataToken, &dst->serverDerivedFromDataToken); \
_mongocrypt_buffer_copy_to(&src->encryptedTokens, &dst->encryptedTokens); \
}

DEF_TEXT_SEARCH_TOKEN_SET_INIT_CLEANUP(Exact)
Expand Down
8 changes: 4 additions & 4 deletions src/mongocrypt-buffer.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -349,23 +349,23 @@ static void _mongocrypt_buffer_copy_as_bson_value(_mongocrypt_buffer_t *plaintex
bson_destroy(&wrapper);
}

bool append_iter(bson_t *bson, const void *iter, int len) {
static bool _append_iter(bson_t *bson, const void *iter, int len) {
return bson_append_iter(bson, "", 0, (const bson_iter_t *)iter);
}

bool append_utf8(bson_t *bson, const void *str, int len) {
static bool _append_utf8(bson_t *bson, const void *str, int len) {
return bson_append_utf8(bson, "", 0, (const char *)str, len);
}

void _mongocrypt_buffer_copy_from_string_as_bson_value(_mongocrypt_buffer_t *plaintext, const char *str, int len) {
BSON_ASSERT_PARAM(str);
BSON_ASSERT(len >= 0);
_mongocrypt_buffer_copy_as_bson_value(plaintext, append_utf8, str, len);
_mongocrypt_buffer_copy_as_bson_value(plaintext, _append_utf8, str, len);
}

void _mongocrypt_buffer_from_iter(_mongocrypt_buffer_t *plaintext, bson_iter_t *iter) {
BSON_ASSERT_PARAM(iter);
_mongocrypt_buffer_copy_as_bson_value(plaintext, append_iter, iter, 0);
_mongocrypt_buffer_copy_as_bson_value(plaintext, _append_iter, iter, 0);
}

bool _mongocrypt_buffer_from_uuid_iter(_mongocrypt_buffer_t *buf, bson_iter_t *iter) {
Expand Down
147 changes: 84 additions & 63 deletions src/mongocrypt-marking.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1192,29 +1192,36 @@ static bool _fle2_generate_TextSearchTokenSets(_mongocrypt_key_broker_t *kb,
mc_substring_set_iter_init(&set_itr, encodeSets->substring_set);

while (mc_substring_set_iter_next(&set_itr, &substring, &bytelen, &appendCount)) {
for (; appendCount > 0; appendCount--) {
mc_TextSubstringTokenSet_t tset = {{0}};
mc_TextSubstringTokenSet_init(&tset);

_mongocrypt_buffer_t asBsonValue;
_mongocrypt_buffer_init(&asBsonValue);
BSON_ASSERT(bytelen < INT_MAX);
_mongocrypt_buffer_copy_from_string_as_bson_value(&asBsonValue, substring, (int)bytelen);

if (!_fle2_generate_TextSubstringTokenSet(kb,
&tset,
&asBsonValue,
contentionFactor,
common.collectionsLevel1Token,
common.serverTokenDerivationLevel1Token,
status)) {
_mongocrypt_buffer_cleanup(&asBsonValue);
mc_TextSubstringTokenSet_cleanup(&tset);
goto fail;
}
_mc_array_append_val(&tsts->substringArray, tset); // moves ownership of tset
BSON_ASSERT(appendCount > 0);
BSON_ASSERT(bytelen < INT_MAX);

mc_TextSubstringTokenSet_t tset = {{0}};
mc_TextSubstringTokenSet_init(&tset);

_mongocrypt_buffer_t asBsonValue;
_mongocrypt_buffer_init(&asBsonValue);
_mongocrypt_buffer_copy_from_string_as_bson_value(&asBsonValue, substring, (int)bytelen);

if (!_fle2_generate_TextSubstringTokenSet(kb,
&tset,
&asBsonValue,
contentionFactor,
common.collectionsLevel1Token,
common.serverTokenDerivationLevel1Token,
status)) {
_mongocrypt_buffer_cleanup(&asBsonValue);
mc_TextSubstringTokenSet_cleanup(&tset);
goto fail;
}
_mongocrypt_buffer_cleanup(&asBsonValue);

for (; appendCount > 1; appendCount--) {
mc_TextSubstringTokenSet_t tset_copy;
mc_TextSubstringTokenSet_init(&tset_copy);
mc_TextSubstringTokenSet_copy(&tset, &tset_copy);
_mc_array_append_val(&tsts->substringArray, tset_copy); // array now owns tset_copy
}
_mc_array_append_val(&tsts->substringArray, tset); // array now owns tset
}
}

Expand All @@ -1224,29 +1231,36 @@ static bool _fle2_generate_TextSearchTokenSets(_mongocrypt_key_broker_t *kb,
mc_affix_set_iter_init(&set_itr, encodeSets->suffix_set);

while (mc_affix_set_iter_next(&set_itr, &substring, &bytelen, &appendCount)) {
for (; appendCount > 0; appendCount--) {
mc_TextSuffixTokenSet_t tset = {{0}};
mc_TextSuffixTokenSet_init(&tset);

_mongocrypt_buffer_t asBsonValue;
_mongocrypt_buffer_init(&asBsonValue);
BSON_ASSERT(bytelen < INT_MAX);
_mongocrypt_buffer_copy_from_string_as_bson_value(&asBsonValue, substring, (int)bytelen);

if (!_fle2_generate_TextSuffixTokenSet(kb,
&tset,
&asBsonValue,
contentionFactor,
common.collectionsLevel1Token,
common.serverTokenDerivationLevel1Token,
status)) {
_mongocrypt_buffer_cleanup(&asBsonValue);
mc_TextSuffixTokenSet_cleanup(&tset);
goto fail;
}
_mc_array_append_val(&tsts->suffixArray, tset); // moves ownership of tset
BSON_ASSERT(appendCount > 0);
BSON_ASSERT(bytelen < INT_MAX);

mc_TextSuffixTokenSet_t tset = {{0}};
mc_TextSuffixTokenSet_init(&tset);

_mongocrypt_buffer_t asBsonValue;
_mongocrypt_buffer_init(&asBsonValue);
_mongocrypt_buffer_copy_from_string_as_bson_value(&asBsonValue, substring, (int)bytelen);

if (!_fle2_generate_TextSuffixTokenSet(kb,
&tset,
&asBsonValue,
contentionFactor,
common.collectionsLevel1Token,
common.serverTokenDerivationLevel1Token,
status)) {
_mongocrypt_buffer_cleanup(&asBsonValue);
mc_TextSuffixTokenSet_cleanup(&tset);
goto fail;
}
_mongocrypt_buffer_cleanup(&asBsonValue);

for (; appendCount > 1; appendCount--) {
mc_TextSuffixTokenSet_t tset_copy;
mc_TextSuffixTokenSet_init(&tset_copy);
mc_TextSuffixTokenSet_copy(&tset, &tset_copy);
_mc_array_append_val(&tsts->suffixArray, tset_copy); // array now owns tset_copy
}
_mc_array_append_val(&tsts->suffixArray, tset); // array now owns tset
}
}

Expand All @@ -1256,29 +1270,36 @@ static bool _fle2_generate_TextSearchTokenSets(_mongocrypt_key_broker_t *kb,
mc_affix_set_iter_init(&set_itr, encodeSets->prefix_set);

while (mc_affix_set_iter_next(&set_itr, &substring, &bytelen, &appendCount)) {
for (; appendCount > 0; appendCount--) {
mc_TextPrefixTokenSet_t tset = {{0}};
mc_TextPrefixTokenSet_init(&tset);

_mongocrypt_buffer_t asBsonValue;
_mongocrypt_buffer_init(&asBsonValue);
BSON_ASSERT(bytelen < INT_MAX);
_mongocrypt_buffer_copy_from_string_as_bson_value(&asBsonValue, substring, (int)bytelen);

if (!_fle2_generate_TextPrefixTokenSet(kb,
&tset,
&asBsonValue,
contentionFactor,
common.collectionsLevel1Token,
common.serverTokenDerivationLevel1Token,
status)) {
_mongocrypt_buffer_cleanup(&asBsonValue);
mc_TextPrefixTokenSet_cleanup(&tset);
goto fail;
}
_mc_array_append_val(&tsts->prefixArray, tset); // moves ownership of tset
BSON_ASSERT(appendCount > 0);
BSON_ASSERT(bytelen < INT_MAX);

mc_TextPrefixTokenSet_t tset = {{0}};
mc_TextPrefixTokenSet_init(&tset);

_mongocrypt_buffer_t asBsonValue;
_mongocrypt_buffer_init(&asBsonValue);
_mongocrypt_buffer_copy_from_string_as_bson_value(&asBsonValue, substring, (int)bytelen);

if (!_fle2_generate_TextPrefixTokenSet(kb,
&tset,
&asBsonValue,
contentionFactor,
common.collectionsLevel1Token,
common.serverTokenDerivationLevel1Token,
status)) {
_mongocrypt_buffer_cleanup(&asBsonValue);
mc_TextPrefixTokenSet_cleanup(&tset);
goto fail;
}
_mongocrypt_buffer_cleanup(&asBsonValue);

for (; appendCount > 1; appendCount--) {
mc_TextPrefixTokenSet_t tset_copy;
mc_TextPrefixTokenSet_init(&tset_copy);
mc_TextPrefixTokenSet_copy(&tset, &tset_copy);
_mc_array_append_val(&tsts->prefixArray, tset_copy); // array now owns tset_copy
}
_mc_array_append_val(&tsts->prefixArray, tset); // moves ownership of tset
}
}
payload->textSearchTokenSets.set = true;
Expand Down
6 changes: 3 additions & 3 deletions test/test-mongocrypt-marking.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1345,15 +1345,15 @@ static void validate_text_search_ciphertext(_mongocrypt_tester_t *tester,
size_t tscount = 0;
ASSERT(bson_iter_init_find(&b_iter, &b_bson, "s"));
tscount = validate_text_search_token_set_array_common(&b_iter, crypt);
ASSERT(expected_tag_counts.substrings == tscount);
ASSERT_CMPSIZE_T(expected_tag_counts.substrings, ==, tscount);

ASSERT(bson_iter_init_find(&b_iter, &b_bson, "u"));
tscount = validate_text_search_token_set_array_common(&b_iter, crypt);
ASSERT(expected_tag_counts.suffixes == tscount);
ASSERT_CMPSIZE_T(expected_tag_counts.suffixes, ==, tscount);

ASSERT(bson_iter_init_find(&b_iter, &b_bson, "p"));
tscount = validate_text_search_token_set_array_common(&b_iter, crypt);
ASSERT(expected_tag_counts.prefixes == tscount);
ASSERT_CMPSIZE_T(expected_tag_counts.prefixes, ==, tscount);
}

mc_ServerDataEncryptionLevel1Token_destroy(sdel1Token);
Expand Down