diff options
author | Eric Wong <e@yhbt.net> | 2020-09-19 09:37:08 +0000 |
---|---|---|
committer | Eric Wong <e@80x24.org> | 2020-09-19 21:39:42 +0000 |
commit | e2f85d6bda87a8c6b25cc340b569ba0a20c2a1eb (patch) | |
tree | 9ba47b5762b7a93a6e94367cbccdb033ed637256 /lib/PublicInbox/gcf2_libgit2.h | |
parent | 05fe38843c2e13cd0c368f8dd7501e4e57c3a829 (diff) | |
download | public-inbox-e2f85d6bda87a8c6b25cc340b569ba0a20c2a1eb.tar.gz |
Having tens of thousands of inboxes and associated git processes won't work well, so we'll use libgit2 to access the object DB directly. We only care about OID lookups and won't need to rely on per-repo revision names or paths. The Git::Raw XS package won't be used since its manpages don't promise a stable API. Since we already use Inline::C and have experience with I::C when it comes to compatibility, this only introduces libgit2 itself as a source of new incompatibilities. This also provides an excuse for me to writev(2) to reduce syscalls, but liburing is on the horizon for next year.
Diffstat (limited to 'lib/PublicInbox/gcf2_libgit2.h')
-rw-r--r-- | lib/PublicInbox/gcf2_libgit2.h | 139 |
1 files changed, 139 insertions, 0 deletions
diff --git a/lib/PublicInbox/gcf2_libgit2.h b/lib/PublicInbox/gcf2_libgit2.h new file mode 100644 index 00000000..d9c79cf9 --- /dev/null +++ b/lib/PublicInbox/gcf2_libgit2.h @@ -0,0 +1,139 @@ +/* + * Copyright (C) 2020 all contributors <meta@public-inbox.org> + * License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt> + * + * libgit2 for Inline::C + * Avoiding Git::Raw since it doesn't guarantee a stable API, + * while libgit2 itself seems reasonably stable. + */ +#include <git2.h> +#include <sys/uio.h> +#include <errno.h> +#include <poll.h> + +static void croak_if_err(int rc, const char *msg) +{ + if (rc != GIT_OK) { + const git_error *e = giterr_last(); + + croak("%d %s (%s)", rc, msg, e ? e->message : "unknown"); + } +} + +SV *new() +{ + git_odb *odb; + SV *ref, *self; + int rc = git_odb_new(&odb); + croak_if_err(rc, "git_odb_new"); + + ref = newSViv((IV)odb); + self = newRV_noinc(ref); + sv_bless(self, gv_stashpv("PublicInbox::Gcf2", GV_ADD)); + SvREADONLY_on(ref); + + return self; +} + +static git_odb *odb_ptr(SV *self) +{ + return (git_odb *)SvIV(SvRV(self)); +} + +void DESTROY(SV *self) +{ + git_odb_free(odb_ptr(self)); +} + +/* needs "$GIT_DIR/objects", not $GIT_DIR */ +void add_alternate(SV *self, const char *objects_path) +{ + int rc = git_odb_add_disk_alternate(odb_ptr(self), objects_path); + croak_if_err(rc, "git_odb_add_disk_alternate"); +} + +/* this requires an unabbreviated git OID */ +#define CAPA(v) (sizeof(v) / sizeof((v)[0])) +void cat_oid(SV *self, int fd, SV *oidsv) +{ + /* + * adjust when libgit2 gets SHA-256 support, we return the + * same header as git-cat-file --batch "$OID $TYPE $SIZE\n" + */ + char hdr[GIT_OID_HEXSZ + sizeof(" commit 18446744073709551615")]; + struct iovec vec[3]; + size_t nvec = CAPA(vec); + git_oid oid; + git_odb_object *object = NULL; + int rc, err = 0; + STRLEN oidlen; + char *oidptr = SvPV(oidsv, oidlen); + + /* same trailer as git-cat-file --batch */ + vec[2].iov_len = 1; + vec[2].iov_base = "\n"; + + rc = git_oid_fromstrn(&oid, oidptr, oidlen); + if (rc == GIT_OK) + rc = git_odb_read(&object, odb_ptr(self), &oid); + if (rc == GIT_OK) { + vec[0].iov_base = hdr; + vec[1].iov_base = (void *)git_odb_object_data(object); + vec[1].iov_len = git_odb_object_size(object); + + git_oid_nfmt(hdr, GIT_OID_HEXSZ, git_odb_object_id(object)); + vec[0].iov_len = GIT_OID_HEXSZ + + snprintf(hdr + GIT_OID_HEXSZ, + sizeof(hdr) - GIT_OID_HEXSZ, + " %s %zu\n", + git_object_type2string( + git_odb_object_type(object)), + vec[1].iov_len); + } else { + vec[0].iov_base = oidptr; + vec[0].iov_len = oidlen; + vec[1].iov_base = " missing"; + vec[1].iov_len = strlen(vec[1].iov_base); + } + while (nvec && !err) { + ssize_t w = writev(fd, vec + CAPA(vec) - nvec, nvec); + + if (w > 0) { + size_t done = 0; + size_t i; + + for (i = CAPA(vec) - nvec; i < CAPA(vec); i++) { + if (w >= vec[i].iov_len) { + /* fully written vec */ + w -= vec[i].iov_len; + done++; + } else { /* partially written vec */ + char *p = vec[i].iov_base; + vec[i].iov_base = p + w; + vec[i].iov_len -= w; + break; + } + } + nvec -= done; + } else if (w < 0) { + err = errno; + switch (err) { + case EAGAIN: { + struct pollfd pfd; + pfd.events = POLLOUT; + pfd.fd = fd; + poll(&pfd, 1, -1); + } + /* fall-through */ + case EINTR: + err = 0; + } + } else { /* w == 0 */ + err = ENOSPC; + } + } + if (object) + git_odb_object_free(object); + if (err) + croak("writev error: %s", strerror(err)); +} |