[PATCH GSoC v16 12/13] cat-file: validate remote atoms with an allow-list
From: Pablo Sabater <hidden>
Date: 2026-07-10 16:41:39
Subsystem:
documentation, the rest · Maintainers:
Jonathan Corbet, Linus Torvalds
strstr() is not enough to validate the format placeholders in remote-object-info causing two errors: 1. Atoms recognized by expand_atom() but the remote doesn't returns 1, but data->type contains garbage causing segfault. 2. expand_atom() returns 0 for unknown atoms, calling strbuf_expand_bad_format() which ends up dying, blocking local queries if the same format is shared. Add an allow-list with the supported atoms at the top of expand_atom(). In remote mode, unsupported atoms return 1 leaving the buffer empty, honoring how for-each-ref handles known but inapplicable atoms. As extra safety, initialize data->type to OBJ_BAD and add a NULL check for type_name() so uninitialized data doesn't cause segfault. Update tests that expect previous die() behavior to expect an empty string and add an explicit test for empty string return on unknown placeholder. Update cat-file command documentation regarding remote-object-info. Mentored-by: Karthik Nayak [off-list ref] Mentored-by: Chandra Pratap [off-list ref] Signed-off-by: Pablo Sabater <redacted> --- Documentation/git-cat-file.adoc | 2 +- builtin/cat-file.c | 41 +++++++++++++++++++++++++++------- t/t1017-cat-file-remote-object-info.sh | 27 ++++++++++++++++++---- 3 files changed, 57 insertions(+), 13 deletions(-)
diff --git a/Documentation/git-cat-file.adoc b/Documentation/git-cat-file.adoc
index a7fa6674c3..643eac9245 100644
--- a/Documentation/git-cat-file.adoc
+++ b/Documentation/git-cat-file.adoc@@ -451,7 +451,7 @@ CAVEATS Note that since only `%(objectname)` and `%(objectsize)` are currently supported by the `remote-object-info` command. Using any other placeholder in -the format string will raise an error. +the format string will return an empty string in its position. Note that the sizes of objects on disk are reported accurately, but care should be taken in drawing conclusions about which refs or objects are
diff --git a/builtin/cat-file.c b/builtin/cat-file.c
index 77ecccdda3..af388b6238 100644
--- a/builtin/cat-file.c
+++ b/builtin/cat-file.c@@ -333,8 +333,18 @@ struct expand_data { * optimized out. */ unsigned skip_object_info : 1; + + /* + * Flags about when an object info is being fetched from remote. + */ + unsigned is_remote:1; +}; +#define EXPAND_DATA_INIT { .mode = S_IFINVALID, .type = OBJ_BAD } + +static const char *remote_object_info_atoms[] = { + "objectname", + "objectsize", }; -#define EXPAND_DATA_INIT { .mode = S_IFINVALID } static int is_atom(const char *atom, const char *s, int slen) {
@@ -345,14 +355,31 @@ static int is_atom(const char *atom, const char *s, int slen) static int expand_atom(struct strbuf *sb, const char *atom, int len, struct expand_data *data) { + if (data->is_remote) { + size_t i, allowed_nr = ARRAY_SIZE(remote_object_info_atoms); + for (i = 0; i < allowed_nr; i++) + if (is_atom(remote_object_info_atoms[i], atom, len)) + break; + + /* + * On remote, skip unsupported atoms returning an empty sb, + * honoring how for-each-ref handles known but inapplicable + * atoms (e.g. %(tagger)). + */ + if (i == allowed_nr) + return 1; + } + if (is_atom("objectname", atom, len)) { if (!data->mark_query) strbuf_add_oid_hex(sb, &data->oid); } else if (is_atom("objecttype", atom, len)) { - if (data->mark_query) + if (data->mark_query) { data->info.typep = &data->type; - else - strbuf_addstr(sb, type_name(data->type)); + } else { + const char *t = type_name(data->type); + strbuf_addstr(sb, t ? t : ""); + } } else if (is_atom("objectsize", atom, len)) { if (data->mark_query) data->info.sizep = &data->size;
@@ -709,10 +736,6 @@ static int get_remote_info(struct batch_options *opt, CALLOC_ARRAY(*remote_object_info, object_info_oids->nr); gtransport->smart_options->object_info_oids = object_info_oids; - /* 'objectsize' is the only option currently supported */ - if (!strstr(opt->format, "%(objectsize)")) - die(_("%s is currently not supported with remote-object-info"), opt->format); - string_list_append(&object_info_options, "size"); if (object_info_options.nr > 0) {
@@ -842,7 +865,9 @@ static void parse_cmd_remote_object_info(struct batch_options *opt, */ data->size = *remote_object_info[i].sizep; opt->batch_mode = BATCH_MODE_INFO; + data->is_remote = 1; batch_object_write(argv[i + 1], output, opt, data, NULL, 0); + data->is_remote = 0; } else { report_object_status(opt, oid_to_hex(&data->oid), &data->oid, "missing"); }
diff --git a/t/t1017-cat-file-remote-object-info.sh b/t/t1017-cat-file-remote-object-info.sh
index 49b6660934..6bc863c391 100755
--- a/t/t1017-cat-file-remote-object-info.sh
+++ b/t/t1017-cat-file-remote-object-info.sh@@ -236,6 +236,21 @@ test_expect_success 'remote-object-info does not die on missing oid like info' ' ) ' +# This tests depends on %(objecttype) not being supported yet, once supported +# it needs to be updated. +test_expect_success 'unsupported placeholder on remote returns empty string' ' + ( + set_transport_variables "$daemon_parent" && + cd "$daemon_parent/daemon_client_empty" && + + echo "" >expect && + git cat-file --batch-command="%(objecttype)" >actual <<-EOF && + remote-object-info "$GIT_DAEMON_URL/parent" $hello_oid + EOF + test_cmp expect actual + ) +' + # Test --batch-command remote-object-info with 'git://' and # transfer.advertiseobjectinfo set to false, i.e. server does not have object-info capability test_expect_success 'batch-command remote-object-info git:// fails when transfer.advertiseobjectinfo=false' '
@@ -575,10 +590,12 @@ test_expect_success 'remote-object-info fails on unsupported filter option (obje set_transport_variables "$HTTPD_DOCUMENT_ROOT_PATH/http_parent" && cd "$HTTPD_DOCUMENT_ROOT_PATH/http_parent" && - test_must_fail git cat-file --batch-command="%(objectsize:disk)" 2>err <<-EOF && + echo "$hello_oid " >expect && + + git cat-file --batch-command="%(objectname) %(objectsize:disk)" >actual <<-EOF && remote-object-info "$HTTPD_URL/smart/http_parent" $hello_oid EOF - test_grep "%(objectsize:disk) is currently not supported with remote-object-info" err + test_cmp expect actual ) '
@@ -587,10 +604,12 @@ test_expect_success 'remote-object-info fails on unsupported filter option (delt set_transport_variables "$HTTPD_DOCUMENT_ROOT_PATH/http_parent" && cd "$HTTPD_DOCUMENT_ROOT_PATH/http_parent" && - test_must_fail git cat-file --batch-command="%(deltabase)" 2>err <<-EOF && + echo "" >expect && + + git cat-file --batch-command="%(deltabase)" >actual <<-EOF && remote-object-info "$HTTPD_URL/smart/http_parent" $hello_oid EOF - test_grep "%(deltabase) is currently not supported with remote-object-info" err + test_cmp expect actual ) '
--
2.54.0