1
0
Fork 0
mirror of https://github.com/git/git.git synced 2024-05-19 14:06:37 +02:00
git/bundle-uri.c

627 lines
14 KiB
C
Raw Normal View History

#include "cache.h"
#include "bundle-uri.h"
#include "bundle.h"
#include "object-store.h"
#include "refs.h"
#include "run-command.h"
bundle-uri: create bundle_list struct and helpers It will likely be rare where a user uses a single bundle URI and expects that URI to point to a bundle. Instead, that URI will likely be a list of bundles provided in some format. Alternatively, the Git server could advertise a list of bundles. In anticipation of these two ways of advertising multiple bundles, create a data structure that represents such a list. This will be populated using a common API, but for now focus on what data can be represented. Each list contains a number of remote_bundle_info structs. These contain an 'id' that is used to uniquely identify them in the list, and also a 'uri' that contains the location of its data. Finally, there is a strbuf containing the filename used when Git downloads the contents to disk. The list itself stores these remote_bundle_info structs in a hashtable using 'id' as the key. The order of the structs in the input is considered unimportant, but future modifications to the format and these data structures will place ordering possibilities on the set. The list also has a few "global" properties, including the version (used when parsing the list) and the mode. The mode is one of these two options: 1. BUNDLE_MODE_ALL: all listed URIs are intended to be combined together. The client should download all of the advertised data to have a complete copy of the data. 2. BUNDLE_MODE_ANY: any one listed item is sufficient to have a complete copy of the data. The client can choose arbitrarily from these options. In the future, the client may use pings to find the closest URI among geodistributed replicas, or use some other heuristic information added to the format. This API is currently unused, but will soon be expanded with parsing logic and then be consumed by the bundle URI download logic. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:29 +02:00
#include "hashmap.h"
#include "pkt-line.h"
bundle-uri: create base key-value pair parsing There will be two primary ways to advertise a bundle list: as a list of packet lines in Git's protocol v2 and as a config file served from a bundle URI. Both of these fundamentally use a list of key-value pairs. We will use the same set of key-value pairs across these formats. Create a new bundle_list_update() method that is currently unusued, but will be used in the next change. It inspects each key to see if it is understood and then applies it to the given bundle_list. Here are the keys that we teach Git to understand: * bundle.version: This value should be an integer. Git currently understands only version 1 and will ignore the list if the version is any other value. This version can be increased in the future if we need to add new keys that Git should not ignore. We can add new "heuristic" keys without incrementing the version. * bundle.mode: This value should be one of "all" or "any". If this mode is not understood, then Git will ignore the list. This mode indicates whether Git needs all of the bundle list items to make a complete view of the content or if any single item is sufficient. The rest of the keys use a bundle identifier "<id>" as part of the key name. Keys using the same "<id>" describe a single bundle list item. * bundle.<id>.uri: This stores the URI of the bundle item. This currently is expected to be an absolute URI, but will be relaxed to be a relative URI in the future. While parsing, return an error if a URI key is repeated, since we can make that restriction with bundle lists. Make the git_parse_int() method global so we can parse the integer version value carefully. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:30 +02:00
#include "config.h"
bundle-uri: create bundle_list struct and helpers It will likely be rare where a user uses a single bundle URI and expects that URI to point to a bundle. Instead, that URI will likely be a list of bundles provided in some format. Alternatively, the Git server could advertise a list of bundles. In anticipation of these two ways of advertising multiple bundles, create a data structure that represents such a list. This will be populated using a common API, but for now focus on what data can be represented. Each list contains a number of remote_bundle_info structs. These contain an 'id' that is used to uniquely identify them in the list, and also a 'uri' that contains the location of its data. Finally, there is a strbuf containing the filename used when Git downloads the contents to disk. The list itself stores these remote_bundle_info structs in a hashtable using 'id' as the key. The order of the structs in the input is considered unimportant, but future modifications to the format and these data structures will place ordering possibilities on the set. The list also has a few "global" properties, including the version (used when parsing the list) and the mode. The mode is one of these two options: 1. BUNDLE_MODE_ALL: all listed URIs are intended to be combined together. The client should download all of the advertised data to have a complete copy of the data. 2. BUNDLE_MODE_ANY: any one listed item is sufficient to have a complete copy of the data. The client can choose arbitrarily from these options. In the future, the client may use pings to find the closest URI among geodistributed replicas, or use some other heuristic information added to the format. This API is currently unused, but will soon be expanded with parsing logic and then be consumed by the bundle URI download logic. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:29 +02:00
static int compare_bundles(const void *hashmap_cmp_fn_data,
const struct hashmap_entry *he1,
const struct hashmap_entry *he2,
const void *id)
{
const struct remote_bundle_info *e1 =
container_of(he1, const struct remote_bundle_info, ent);
const struct remote_bundle_info *e2 =
container_of(he2, const struct remote_bundle_info, ent);
return strcmp(e1->id, id ? (const char *)id : e2->id);
}
void init_bundle_list(struct bundle_list *list)
{
memset(list, 0, sizeof(*list));
/* Implied defaults. */
list->mode = BUNDLE_MODE_ALL;
list->version = 1;
hashmap_init(&list->bundles, compare_bundles, NULL, 0);
}
static int clear_remote_bundle_info(struct remote_bundle_info *bundle,
void *data)
{
FREE_AND_NULL(bundle->id);
FREE_AND_NULL(bundle->uri);
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
FREE_AND_NULL(bundle->file);
bundle->unbundled = 0;
bundle-uri: create bundle_list struct and helpers It will likely be rare where a user uses a single bundle URI and expects that URI to point to a bundle. Instead, that URI will likely be a list of bundles provided in some format. Alternatively, the Git server could advertise a list of bundles. In anticipation of these two ways of advertising multiple bundles, create a data structure that represents such a list. This will be populated using a common API, but for now focus on what data can be represented. Each list contains a number of remote_bundle_info structs. These contain an 'id' that is used to uniquely identify them in the list, and also a 'uri' that contains the location of its data. Finally, there is a strbuf containing the filename used when Git downloads the contents to disk. The list itself stores these remote_bundle_info structs in a hashtable using 'id' as the key. The order of the structs in the input is considered unimportant, but future modifications to the format and these data structures will place ordering possibilities on the set. The list also has a few "global" properties, including the version (used when parsing the list) and the mode. The mode is one of these two options: 1. BUNDLE_MODE_ALL: all listed URIs are intended to be combined together. The client should download all of the advertised data to have a complete copy of the data. 2. BUNDLE_MODE_ANY: any one listed item is sufficient to have a complete copy of the data. The client can choose arbitrarily from these options. In the future, the client may use pings to find the closest URI among geodistributed replicas, or use some other heuristic information added to the format. This API is currently unused, but will soon be expanded with parsing logic and then be consumed by the bundle URI download logic. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:29 +02:00
return 0;
}
void clear_bundle_list(struct bundle_list *list)
{
if (!list)
return;
for_all_bundles_in_list(list, clear_remote_bundle_info, NULL);
hashmap_clear_and_free(&list->bundles, struct remote_bundle_info, ent);
}
int for_all_bundles_in_list(struct bundle_list *list,
bundle_iterator iter,
void *data)
{
struct remote_bundle_info *info;
struct hashmap_iter i;
hashmap_for_each_entry(&list->bundles, &i, info, ent) {
int result = iter(info, data);
if (result)
return result;
}
return 0;
}
static int summarize_bundle(struct remote_bundle_info *info, void *data)
{
FILE *fp = data;
fprintf(fp, "[bundle \"%s\"]\n", info->id);
fprintf(fp, "\turi = %s\n", info->uri);
return 0;
}
void print_bundle_list(FILE *fp, struct bundle_list *list)
{
const char *mode;
switch (list->mode) {
case BUNDLE_MODE_ALL:
mode = "all";
break;
case BUNDLE_MODE_ANY:
mode = "any";
break;
case BUNDLE_MODE_NONE:
default:
mode = "<unknown>";
}
fprintf(fp, "[bundle]\n");
fprintf(fp, "\tversion = %d\n", list->version);
fprintf(fp, "\tmode = %s\n", mode);
for_all_bundles_in_list(list, summarize_bundle, fp);
}
bundle-uri: create base key-value pair parsing There will be two primary ways to advertise a bundle list: as a list of packet lines in Git's protocol v2 and as a config file served from a bundle URI. Both of these fundamentally use a list of key-value pairs. We will use the same set of key-value pairs across these formats. Create a new bundle_list_update() method that is currently unusued, but will be used in the next change. It inspects each key to see if it is understood and then applies it to the given bundle_list. Here are the keys that we teach Git to understand: * bundle.version: This value should be an integer. Git currently understands only version 1 and will ignore the list if the version is any other value. This version can be increased in the future if we need to add new keys that Git should not ignore. We can add new "heuristic" keys without incrementing the version. * bundle.mode: This value should be one of "all" or "any". If this mode is not understood, then Git will ignore the list. This mode indicates whether Git needs all of the bundle list items to make a complete view of the content or if any single item is sufficient. The rest of the keys use a bundle identifier "<id>" as part of the key name. Keys using the same "<id>" describe a single bundle list item. * bundle.<id>.uri: This stores the URI of the bundle item. This currently is expected to be an absolute URI, but will be relaxed to be a relative URI in the future. While parsing, return an error if a URI key is repeated, since we can make that restriction with bundle lists. Make the git_parse_int() method global so we can parse the integer version value carefully. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:30 +02:00
/**
* Given a key-value pair, update the state of the given bundle list.
* Returns 0 if the key-value pair is understood. Returns -1 if the key
* is not understood or the value is malformed.
*/
static int bundle_list_update(const char *key, const char *value,
struct bundle_list *list)
{
struct strbuf id = STRBUF_INIT;
struct remote_bundle_info lookup = REMOTE_BUNDLE_INFO_INIT;
struct remote_bundle_info *bundle;
const char *subsection, *subkey;
size_t subsection_len;
if (parse_config_key(key, "bundle", &subsection, &subsection_len, &subkey))
return -1;
if (!subsection_len) {
if (!strcmp(subkey, "version")) {
int version;
if (!git_parse_int(value, &version))
return -1;
if (version != 1)
return -1;
list->version = version;
return 0;
}
if (!strcmp(subkey, "mode")) {
if (!strcmp(value, "all"))
list->mode = BUNDLE_MODE_ALL;
else if (!strcmp(value, "any"))
list->mode = BUNDLE_MODE_ANY;
else
return -1;
return 0;
}
/* Ignore other unknown global keys. */
return 0;
}
strbuf_add(&id, subsection, subsection_len);
/*
* Check for an existing bundle with this <id>, or create one
* if necessary.
*/
lookup.id = id.buf;
hashmap_entry_init(&lookup.ent, strhash(lookup.id));
if (!(bundle = hashmap_get_entry(&list->bundles, &lookup, ent, NULL))) {
CALLOC_ARRAY(bundle, 1);
bundle->id = strbuf_detach(&id, NULL);
hashmap_entry_init(&bundle->ent, strhash(bundle->id));
hashmap_add(&list->bundles, &bundle->ent);
}
strbuf_release(&id);
if (!strcmp(subkey, "uri")) {
if (bundle->uri)
return -1;
bundle->uri = xstrdup(value);
return 0;
}
/*
* At this point, we ignore any information that we don't
* understand, assuming it to be hints for a heuristic the client
* does not currently understand.
*/
return 0;
}
bundle-uri: parse bundle list in config format When a bundle provider wants to operate independently from a Git remote, they want to provide a single, consistent URI that users can use in their 'git clone --bundle-uri' commands. At this point, the Git client expects that URI to be a single bundle that can be unbundled and used to bootstrap the rest of the clone from the Git server. This single bundle cannot be re-used to assist with future incremental fetches. To allow for the incremental fetch case, teach Git to understand a bundle list that could be advertised at an independent bundle URI. Such a bundle list is likely to be inspected by human readers, even if only by the bundle provider creating the list. For this reason, we can take our expected "key=value" pairs and instead format them using Git config format. Create bundle_uri_parse_config_format() to parse a file in config format and convert that into a 'struct bundle_list' filled with its understanding of the contents. Be careful to use error_action CONFIG_ERROR_ERROR when calling git_config_from_file_with_options() because the default action for git_config_from_file() is to die() on a parsing error. The current warning isn't particularly helpful if it arises to a user, but it will be made more verbose at a higher layer later. Update 'test-tool bundle-uri' to take this config file format as input. It uses a filename instead of stdin because there is no existing way to parse a FILE pointer in the config machinery. Using git_config_from_mem() is overly complicated and more likely to introduce bugs than this simpler version. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:33 +02:00
static int config_to_bundle_list(const char *key, const char *value, void *data)
{
struct bundle_list *list = data;
return bundle_list_update(key, value, list);
}
int bundle_uri_parse_config_format(const char *uri,
const char *filename,
struct bundle_list *list)
{
int result;
struct config_options opts = {
.error_action = CONFIG_ERROR_ERROR,
};
result = git_config_from_file_with_options(config_to_bundle_list,
filename, list,
&opts);
if (!result && list->mode == BUNDLE_MODE_NONE) {
warning(_("bundle list at '%s' has no mode"), uri);
result = 1;
}
return result;
}
static char *find_temp_filename(void)
{
int fd;
struct strbuf name = STRBUF_INIT;
/*
* Find a temporary filename that is available. This is briefly
* racy, but unlikely to collide.
*/
fd = odb_mkstemp(&name, "bundles/tmp_uri_XXXXXX");
if (fd < 0) {
warning(_("failed to create temporary file"));
return NULL;
}
close(fd);
unlink(name.buf);
return strbuf_detach(&name, NULL);
}
static int download_https_uri_to_file(const char *file, const char *uri)
{
int result = 0;
struct child_process cp = CHILD_PROCESS_INIT;
FILE *child_in = NULL, *child_out = NULL;
struct strbuf line = STRBUF_INIT;
int found_get = 0;
strvec_pushl(&cp.args, "git-remote-https", uri, NULL);
cp.err = -1;
cp.in = -1;
cp.out = -1;
if (start_command(&cp))
return 1;
child_in = fdopen(cp.in, "w");
if (!child_in) {
result = 1;
goto cleanup;
}
child_out = fdopen(cp.out, "r");
if (!child_out) {
result = 1;
goto cleanup;
}
fprintf(child_in, "capabilities\n");
fflush(child_in);
while (!strbuf_getline(&line, child_out)) {
if (!line.len)
break;
if (!strcmp(line.buf, "get"))
found_get = 1;
}
strbuf_release(&line);
if (!found_get) {
result = error(_("insufficient capabilities"));
goto cleanup;
}
fprintf(child_in, "get %s %s\n\n", uri, file);
cleanup:
if (child_in)
fclose(child_in);
if (finish_command(&cp))
return 1;
if (child_out)
fclose(child_out);
return result;
}
static int copy_uri_to_file(const char *filename, const char *uri)
{
const char *out;
if (starts_with(uri, "https:") ||
starts_with(uri, "http:"))
return download_https_uri_to_file(filename, uri);
if (skip_prefix(uri, "file://", &out))
uri = out;
/* Copy as a file */
return copy_file(filename, uri, 0);
}
static int unbundle_from_file(struct repository *r, const char *file)
{
int result = 0;
int bundle_fd;
struct bundle_header header = BUNDLE_HEADER_INIT;
struct string_list_item *refname;
struct strbuf bundle_ref = STRBUF_INIT;
size_t bundle_prefix_len;
if ((bundle_fd = read_bundle_header(file, &header)) < 0)
return 1;
/*
* Skip the reachability walk here, since we will be adding
* a reachable ref pointing to the new tips, which will reach
* the prerequisite commits.
*/
if ((result = unbundle(r, &header, bundle_fd, NULL,
VERIFY_BUNDLE_QUIET)))
return 1;
/*
* Convert all refs/heads/ from the bundle into refs/bundles/
* in the local repository.
*/
strbuf_addstr(&bundle_ref, "refs/bundles/");
bundle_prefix_len = bundle_ref.len;
for_each_string_list_item(refname, &header.references) {
struct object_id *oid = refname->util;
struct object_id old_oid;
const char *branch_name;
int has_old;
if (!skip_prefix(refname->string, "refs/heads/", &branch_name))
continue;
strbuf_setlen(&bundle_ref, bundle_prefix_len);
strbuf_addstr(&bundle_ref, branch_name);
has_old = !read_ref(bundle_ref.buf, &old_oid);
update_ref("fetched bundle", bundle_ref.buf, oid,
has_old ? &old_oid : NULL,
REF_SKIP_OID_VERIFICATION,
UPDATE_REFS_MSG_ON_ERR);
}
bundle_header_release(&header);
return result;
}
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
struct bundle_list_context {
struct repository *r;
struct bundle_list *list;
enum bundle_list_mode mode;
int count;
int depth;
};
/*
* This early definition is necessary because we use indirect recursion:
*
* While iterating through a bundle list that was downloaded as part
* of fetch_bundle_uri_internal(), iterator methods eventually call it
* again, but with depth + 1.
*/
static int fetch_bundle_uri_internal(struct repository *r,
struct remote_bundle_info *bundle,
int depth,
struct bundle_list *list);
static int download_bundle_to_file(struct remote_bundle_info *bundle, void *data)
{
int res;
struct bundle_list_context *ctx = data;
if (ctx->mode == BUNDLE_MODE_ANY && ctx->count)
return 0;
res = fetch_bundle_uri_internal(ctx->r, bundle, ctx->depth + 1, ctx->list);
/*
* Only increment count if the download succeeded. If our mode is
* BUNDLE_MODE_ANY, then we will want to try other URIs in the
* list in case they work instead.
*/
if (!res)
ctx->count++;
/*
* To be opportunistic as possible, we continue iterating and
* download as many bundles as we can, so we can apply the ones
* that work, even in BUNDLE_MODE_ALL mode.
*/
return 0;
}
static int download_bundle_list(struct repository *r,
struct bundle_list *local_list,
struct bundle_list *global_list,
int depth)
{
struct bundle_list_context ctx = {
.r = r,
.list = global_list,
.depth = depth + 1,
.mode = local_list->mode,
};
return for_all_bundles_in_list(local_list, download_bundle_to_file, &ctx);
}
static int fetch_bundle_list_in_config_format(struct repository *r,
struct bundle_list *global_list,
struct remote_bundle_info *bundle,
int depth)
{
int result;
struct bundle_list list_from_bundle;
init_bundle_list(&list_from_bundle);
if ((result = bundle_uri_parse_config_format(bundle->uri,
bundle->file,
&list_from_bundle)))
goto cleanup;
if (list_from_bundle.mode == BUNDLE_MODE_NONE) {
warning(_("unrecognized bundle mode from URI '%s'"),
bundle->uri);
result = -1;
goto cleanup;
}
if ((result = download_bundle_list(r, &list_from_bundle,
global_list, depth)))
goto cleanup;
cleanup:
clear_bundle_list(&list_from_bundle);
return result;
}
/**
* This limits the recursion on fetch_bundle_uri_internal() when following
* bundle lists.
*/
static int max_bundle_uri_depth = 4;
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
/**
* Recursively download all bundles advertised at the given URI
* to files. If the file is a bundle, then add it to the given
* 'list'. Otherwise, expect a bundle list and recurse on the
* URIs in that list according to the list mode (ANY or ALL).
*/
static int fetch_bundle_uri_internal(struct repository *r,
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
struct remote_bundle_info *bundle,
int depth,
struct bundle_list *list)
{
int result = 0;
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
struct remote_bundle_info *bcopy;
if (depth >= max_bundle_uri_depth) {
warning(_("exceeded bundle URI recursion limit (%d)"),
max_bundle_uri_depth);
return -1;
}
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
if (!bundle->file &&
!(bundle->file = find_temp_filename())) {
result = -1;
goto cleanup;
}
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
if ((result = copy_uri_to_file(bundle->file, bundle->uri))) {
warning(_("failed to download bundle from URI '%s'"), bundle->uri);
goto cleanup;
}
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
if ((result = !is_bundle(bundle->file, 1))) {
result = fetch_bundle_list_in_config_format(
r, list, bundle, depth);
if (result)
warning(_("file at URI '%s' is not a bundle or bundle list"),
bundle->uri);
goto cleanup;
}
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
/* Copy the bundle and insert it into the global list. */
CALLOC_ARRAY(bcopy, 1);
bcopy->id = xstrdup(bundle->id);
bcopy->file = xstrdup(bundle->file);
hashmap_entry_init(&bcopy->ent, strhash(bcopy->id));
hashmap_add(&list->bundles, &bcopy->ent);
cleanup:
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
if (result && bundle->file)
unlink(bundle->file);
return result;
}
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
/**
* This loop iterator breaks the loop with nonzero return code on the
* first successful unbundling of a bundle.
*/
static int attempt_unbundle(struct remote_bundle_info *info, void *data)
{
struct repository *r = data;
if (!info->file || info->unbundled)
return 0;
if (!unbundle_from_file(r, info->file)) {
info->unbundled = 1;
return 1;
}
return 0;
}
static int unbundle_all_bundles(struct repository *r,
struct bundle_list *list)
{
/*
* Iterate through all bundles looking for ones that can
* successfully unbundle. If any succeed, then perhaps another
* will succeed in the next attempt.
*
* Keep in mind that a non-zero result for the loop here means
* the loop terminated early on a successful unbundling, which
* signals that we can try again.
*/
while (for_all_bundles_in_list(list, attempt_unbundle, r)) ;
return 0;
}
static int unlink_bundle(struct remote_bundle_info *info, void *data)
{
if (info->file)
unlink_or_warn(info->file);
return 0;
}
int fetch_bundle_uri(struct repository *r, const char *uri)
{
bundle-uri: fetch a list of bundles When the content at a given bundle URI is not understood as a bundle (based on inspecting the initial content), then Git currently gives up and ignores that content. Independent bundle providers may want to split up the bundle content into multiple bundles, but still make them available from a single URI. Teach Git to attempt parsing the bundle URI content as a Git config file providing the key=value pairs for a bundle list. Git then looks at the mode of the list to see if ANY single bundle is sufficient or if ALL bundles are required. The content at the selected URIs are downloaded and the content is inspected again, creating a recursive process. To guard the recursion against malformed or malicious content, limit the recursion depth to a reasonable four for now. This can be converted to a configured value in the future if necessary. The value of four is twice as high as expected to be useful (a bundle list is unlikely to point to more bundle lists). To test this scenario, create an interesting bundle topology where three incremental bundles are built on top of a single full bundle. By using a merge commit, the two middle bundles are "independent" in that they do not require each other in order to unbundle themselves. They each only need the base bundle. The bundle containing the merge commit requires both of the middle bundles, though. This leads to some interesting decisions when unbundling, especially when we later implement heuristics that promote downloading bundles until the prerequisite commits are satisfied. Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-10-12 14:52:36 +02:00
int result;
struct bundle_list list;
struct remote_bundle_info bundle = {
.uri = xstrdup(uri),
.id = xstrdup(""),
};
init_bundle_list(&list);
/* If a bundle is added to this global list, then it is required. */
list.mode = BUNDLE_MODE_ALL;
if ((result = fetch_bundle_uri_internal(r, &bundle, 0, &list)))
goto cleanup;
result = unbundle_all_bundles(r, &list);
cleanup:
for_all_bundles_in_list(&list, unlink_bundle, NULL);
clear_bundle_list(&list);
clear_remote_bundle_info(&bundle, NULL);
return result;
}
protocol v2: add server-side "bundle-uri" skeleton Add a skeleton server-side implementation of a new "bundle-uri" command to protocol v2. This will allow conforming clients to optionally seed their initial clones or incremental fetches from URLs containing "*.bundle" files created with "git bundle create". This change only performs the basic boilerplate of advertising a new protocol v2 capability. The new 'bundle-uri' capability allows a client to request a list of bundles. Right now, the server only returns a flush packet, which corresponds to an empty advertisement. The bundle.* config namespace describes which key-value pairs will be communicated across this interface in future updates. The critical bit right now is that the new boolean uploadPack.adverstiseBundleURIs config value signals whether or not this capability should be advertised at all. An earlier version of this patch [1] used a different transfer format than the "key=value" pairs in the current implementation. The change was made to unify the protocol v2 command with the bundle lists provided by independent bundle servers. Further, the standard allows for the server to advertise a URI that contains a bundle list. This allows users automatically discovering bundle providers that are loosely associated with the origin server, but without the origin server knowing exactly which bundles are currently available. [1] https://lore.kernel.org/git/RFC-patch-v2-01.13-2fc87ce092b-20220311T155841Z-avarab@gmail.com/ The very-deep headings needed to be modified to stop at level 4 due to documentation build issues. These were not recognized in earlier builds since the file was previously in the Documentation/technical/ directory and was built in a different way. With its current location, the heavily-nested details were causing build issues and they are now replaced with a bulletted list of details. Co-authored-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-12-22 16:14:07 +01:00
/**
* API for serve.c.
*/
int bundle_uri_advertise(struct repository *r, struct strbuf *value UNUSED)
{
static int advertise_bundle_uri = -1;
if (advertise_bundle_uri != -1)
goto cached;
advertise_bundle_uri = 0;
repo_config_get_maybe_bool(r, "uploadpack.advertisebundleuris", &advertise_bundle_uri);
cached:
return advertise_bundle_uri;
}
int bundle_uri_command(struct repository *r,
struct packet_reader *request)
{
struct packet_writer writer;
packet_writer_init(&writer, 1);
while (packet_reader_read(request) == PACKET_READ_NORMAL)
die(_("bundle-uri: unexpected argument: '%s'"), request->line);
if (request->status != PACKET_READ_FLUSH)
die(_("bundle-uri: expected flush after arguments"));
/* TODO: Implement the communication */
packet_writer_flush(&writer);
return 0;
}
/**
* General API for {transport,connect}.c etc.
*/
int bundle_uri_parse_line(struct bundle_list *list, const char *line)
{
int result;
const char *equals;
struct strbuf key = STRBUF_INIT;
if (!strlen(line))
return error(_("bundle-uri: got an empty line"));
equals = strchr(line, '=');
if (!equals)
return error(_("bundle-uri: line is not of the form 'key=value'"));
if (line == equals || !*(equals + 1))
return error(_("bundle-uri: line has empty key or value"));
strbuf_add(&key, line, equals - line);
result = bundle_list_update(key.buf, equals + 1, list);
strbuf_release(&key);
return result;
}