blob: 2f5478693e887a8f32d68418fc434271b9585a6d [file]
/* shared.c: global vars + some callback functions
*
* Copyright (C) 2006-2014 cgit Development Team <cgit@lists.zx2c4.com>
*
* Licensed under GNU General Public License v2
* (see COPYING for full license text)
*/
#define USE_THE_REPOSITORY_VARIABLE
#include "cgit.h"
#include <commit-graph.h>
struct cgit_repolist cgit_repolist;
struct cgit_context ctx;
int chk_zero(int result, char *msg)
{
if (result != 0)
die_errno("%s", msg);
return result;
}
int chk_positive(int result, char *msg)
{
if (result <= 0)
die_errno("%s", msg);
return result;
}
int chk_non_negative(int result, char *msg)
{
if (result < 0)
die_errno("%s", msg);
return result;
}
char *cgit_default_repo_desc = "[no description]";
struct cgit_repo *cgit_add_repo(const char *url)
{
struct cgit_repo *ret;
if (++cgit_repolist.count > cgit_repolist.length) {
if (cgit_repolist.length == 0)
cgit_repolist.length = 8;
else
cgit_repolist.length *= 2;
cgit_repolist.repos = xrealloc(cgit_repolist.repos,
cgit_repolist.length *
sizeof(struct cgit_repo));
}
ret = &cgit_repolist.repos[cgit_repolist.count-1];
memset(ret, 0, sizeof(struct cgit_repo));
ret->url = trim_end(url, '/');
*strchrnul(ret->url, '\n') = '\0';
ret->name = ret->url;
ret->path = NULL;
ret->desc = cgit_default_repo_desc;
ret->extra_head_content = NULL;
ret->owner = NULL;
ret->homepage = NULL;
ret->section = ctx.cfg.section;
ret->snapshots = ctx.cfg.snapshots;
ret->enable_blame = ctx.cfg.enable_blame;
ret->enable_commit_graph = ctx.cfg.enable_commit_graph;
ret->enable_follow_links = ctx.cfg.enable_follow_links;
ret->enable_log_filecount = ctx.cfg.enable_log_filecount;
ret->enable_log_linecount = ctx.cfg.enable_log_linecount;
ret->enable_object_reachability_check = ctx.cfg.enable_object_reachability_check;
ret->enable_remote_branches = ctx.cfg.enable_remote_branches;
ret->enable_subject_links = ctx.cfg.enable_subject_links;
ret->enable_html_serving = ctx.cfg.enable_html_serving;
ret->max_stats = ctx.cfg.max_stats;
ret->branch_sort = ctx.cfg.branch_sort;
ret->commit_sort = ctx.cfg.commit_sort;
ret->module_link = ctx.cfg.module_link;
ret->readme = ctx.cfg.readme;
ret->mtime = -1;
ret->about_filter = ctx.cfg.about_filter;
ret->commit_filter = ctx.cfg.commit_filter;
ret->source_filter = ctx.cfg.source_filter;
ret->email_filter = ctx.cfg.email_filter;
ret->owner_filter = ctx.cfg.owner_filter;
ret->clone_url = ctx.cfg.clone_url;
ret->submodules.strdup_strings = 1;
ret->hide = ret->ignore = 0;
return ret;
}
struct cgit_repo *cgit_get_repoinfo(const char *url)
{
int i;
struct cgit_repo *repo;
for (i = 0; i < cgit_repolist.count; i++) {
repo = &cgit_repolist.repos[i];
if (repo->ignore)
continue;
if (!strcmp(repo->url, url))
return repo;
}
return NULL;
}
void cgit_free_commitinfo(struct commitinfo *info)
{
free(info->author);
free(info->author_email);
free(info->committer);
free(info->committer_email);
free(info->subject);
free(info->msg);
free(info->msg_encoding);
free(info);
}
char *trim_end(const char *str, char c)
{
int len;
if (str == NULL)
return NULL;
len = strlen(str);
while (len > 0 && str[len - 1] == c)
len--;
if (len == 0)
return NULL;
return xstrndup(str, len);
}
char *ensure_end(const char *str, char c)
{
size_t len = strlen(str);
char *result;
if (len && str[len - 1] == c)
return xstrndup(str, len);
result = xmalloc(len + 2);
memcpy(result, str, len);
result[len] = '/';
result[len + 1] = '\0';
return result;
}
void strbuf_ensure_end(struct strbuf *sb, char c)
{
if (!sb->len || sb->buf[sb->len - 1] != c)
strbuf_addch(sb, c);
}
void cgit_add_ref(struct reflist *list, struct refinfo *ref)
{
size_t size;
if (list->count >= list->alloc) {
list->alloc += (list->alloc ? list->alloc : 4);
size = list->alloc * sizeof(struct refinfo *);
list->refs = xrealloc(list->refs, size);
}
list->refs[list->count++] = ref;
}
static struct refinfo *cgit_mk_refinfo(const char *refname, const struct object_id *oid)
{
struct refinfo *ref;
ref = xmalloc(sizeof (struct refinfo));
ref->refname = xstrdup(refname);
ref->object = parse_object(the_repository, oid);
switch (ref->object->type) {
case OBJ_TAG:
ref->tag = cgit_parse_tag((struct tag *)ref->object);
break;
case OBJ_COMMIT:
ref->commit = cgit_parse_commit((struct commit *)ref->object);
break;
}
return ref;
}
void cgit_free_taginfo(struct taginfo *tag)
{
if (tag->tagger)
free(tag->tagger);
if (tag->tagger_email)
free(tag->tagger_email);
if (tag->msg)
free(tag->msg);
free(tag);
}
static void cgit_free_refinfo(struct refinfo *ref)
{
if (ref->refname)
free((char *)ref->refname);
switch (ref->object->type) {
case OBJ_TAG:
cgit_free_taginfo(ref->tag);
break;
case OBJ_COMMIT:
cgit_free_commitinfo(ref->commit);
break;
}
free(ref);
}
void cgit_free_reflist_inner(struct reflist *list)
{
int i;
for (i = 0; i < list->count; i++) {
cgit_free_refinfo(list->refs[i]);
}
free(list->refs);
}
int cgit_refs_cb(const struct reference *ref, void *cb_data)
{
struct reflist *list = (struct reflist *)cb_data;
struct refinfo *info = cgit_mk_refinfo(ref->name, ref->oid);
if (info)
cgit_add_ref(list, info);
return 0;
}
struct reachability_tips {
struct repository *repo;
struct commit **commits;
size_t nr;
size_t alloc;
};
static int collect_reachability_tip(const struct reference *ref, void *cb_data)
{
struct reachability_tips *tips = cb_data;
struct commit *commit;
commit = lookup_commit_reference_gently(tips->repo, ref->oid, 1);
if (!commit)
return 0;
ALLOC_GROW(tips->commits, tips->nr + 1, tips->alloc);
tips->commits[tips->nr++] = commit;
return 0;
}
/* Walk the references that 'repo' publishes. The set deliberately mirrors the
* one shown by the refs page, so that an object is never refused here while
* remaining visible there, or the other way around.
*
* 'r' is the object database to read them from and 'repo' the configuration
* that decides which of them count. The two are separate because the caller
* may be asking about a repository other than the one being served.
*/
static void for_each_published_ref(struct repository *r,
const struct cgit_repo *repo,
refs_for_each_cb fn, void *cb_data)
{
struct ref_store *refs = get_main_ref_store(r);
if (refs_for_each_branch_ref(refs, fn, cb_data))
return;
if (refs_for_each_tag_ref(refs, fn, cb_data))
return;
if (repo && repo->enable_remote_branches)
refs_for_each_remote_ref(refs, fn, cb_data);
}
struct ref_target_match {
struct repository *repo;
const struct object_id *oid;
int found;
};
static int match_ref_target(const struct reference *ref, void *cb_data)
{
struct ref_target_match *match = cb_data;
struct object *obj;
if (oideq(ref->oid, match->oid)) {
match->found = 1;
return 1;
}
if (odb_read_object_info(match->repo->objects, ref->oid, NULL) != OBJ_TAG)
return 0;
obj = parse_object(match->repo, ref->oid);
if (!obj)
return 0;
obj = deref_tag(match->repo, obj, NULL, 0);
if (obj && oideq(&obj->oid, match->oid)) {
match->found = 1;
return 1;
}
return 0;
}
/* Determine whether 'oid' is published by 'r', meaning that a reference points
* directly at it or that it is reachable from one. 'repo' supplies the cgit
* configuration governing which references 'r' publishes, and may be NULL.
*
* For an object that is not commit-ish, only the first test is available.
* Asking whether a blob or a tree occurs somewhere in the history would mean
* walking every object in the repository, which no request can afford. The
* objects cgit links to by raw id are exactly the ones a reference points at,
* so the affordable test is also the one that matters: it keeps those links
* working and refuses everything else, which in a shared object database is
* every blob and tree belonging to some other repository.
*
* The reachability half reports the object as published when it cannot answer
* -- an empty repository has no tips to walk from, and an error out of the
* traversal cannot be interpreted -- because refusing to serve a page is the
* more damaging way to be wrong.
*
* Note that this half walks history, which on a large repository costs
* seconds for an old commit even with the generation numbers a commit-graph
* provides. A commit the repository's own commit-graph lists is accepted
* without the walk; see own_commit_graph_lists().
*/
/* Whether 'refname' names one of the references that 'repo' publishes -- the
* same set for_each_published_ref() walks above. A caller that has shown a
* commit to be an ancestor of such a reference has thereby shown it reachable,
* and can skip asking cgit_oid_is_reachable() the broader question.
*
* Not every name that resolves is published: a raw object id resolves, and so
* do refs/notes/ and other references cgit does not show, so this asks where
* the name lands rather than merely whether it resolves.
*/
int cgit_ref_is_published(struct repository *r, const struct cgit_repo *repo,
const char *refname)
{
struct object_id oid;
char *full = NULL;
int published = 0;
if (repo_dwim_ref(r, refname, strlen(refname), &oid, &full, 0) != 1)
goto out;
if (starts_with(full, "refs/heads/") || starts_with(full, "refs/tags/"))
published = 1;
else if (repo && repo->enable_remote_branches &&
starts_with(full, "refs/remotes/"))
published = 1;
out:
free(full);
return published;
}
/* Whether the commit-graph in r's own object directory lists 'commit'.
*
* A graph written with --reachable records what the references reached, so it
* can stand in for the walk; one written from packed objects would vouch for
* every packed commit. cgitrc(5) states that contract and its staleness.
* --reachable also follows references cgit does not publish, such as notes
* and refs/meta/cgit; their commits belong to this repository all the same.
*
* Only a graph layer stored in r's own object directory counts. Git loads a
* graph from an alternate when the repository has none, and a split chain may
* keep its lower layers in one; a shared object store's graph lists the
* commits of every repository borrowing from it, so it vouches for nothing.
*/
static int own_commit_graph_lists(struct repository *r, struct commit *commit)
{
struct commit_graph *g;
uint32_t pos;
g = repo_find_commit_pos_in_graph(r, commit, &pos);
/* Positions run through the whole chain, base layer first. */
while (g && pos < g->num_commits_in_base)
g = g->base_graph;
return g && g->odb_source == r->objects->sources;
}
int cgit_oid_is_reachable(struct repository *r, const struct cgit_repo *repo,
const struct object_id *oid)
{
struct reachability_tips tips = { r, NULL, 0, 0 };
struct ref_target_match match = { r, oid, 0 };
struct commit *commit;
int reachable = 1;
commit = lookup_commit_reference_gently(r, oid, 1);
if (!commit) {
/* An object that is not in the database at all is a different
* question, and one the page handler answers in its own words.
*/
if (!odb_has_object(r->objects, oid, 0))
return 1;
for_each_published_ref(r, repo, match_ref_target, &match);
return match.found;
}
if (own_commit_graph_lists(r, commit))
return 1;
for_each_published_ref(r, repo, collect_reachability_tip, &tips);
if (tips.nr)
reachable = repo_in_merge_bases_many(r, commit, tips.nr,
tips.commits, 1) != 0;
free(tips.commits);
return reachable;
}
void cgit_diff_tree_cb(struct diff_queue_struct *q,
struct diff_options *options, void *data)
{
int i;
for (i = 0; i < q->nr; i++) {
if (q->queue[i]->status == 'U')
continue;
((filepair_fn)data)(q->queue[i]);
}
}
static int load_mmfile(mmfile_t *file, const struct object_id *oid)
{
enum object_type type;
if (is_null_oid(oid)) {
file->ptr = (char *)"";
file->size = 0;
} else {
file->ptr = odb_read_object(the_repository->objects, oid, &type,
(unsigned long *)&file->size);
}
return 1;
}
/*
* Receive diff-buffers from xdiff and concatenate them as
* needed across multiple callbacks.
*
* This is basically a copy of xdiff-interface.c/xdiff_outf(),
* ripped from git and modified to use globals instead of
* a special callback-struct.
*/
static char *diffbuf = NULL;
static int buflen = 0;
static int filediff_cb(void *priv, mmbuffer_t *mb, int nbuf)
{
int i;
for (i = 0; i < nbuf; i++) {
if (mb[i].ptr[mb[i].size-1] != '\n') {
/* Incomplete line */
diffbuf = xrealloc(diffbuf, buflen + mb[i].size);
memcpy(diffbuf + buflen, mb[i].ptr, mb[i].size);
buflen += mb[i].size;
continue;
}
/* we have a complete line */
if (!diffbuf) {
((linediff_fn)priv)(mb[i].ptr, mb[i].size);
continue;
}
diffbuf = xrealloc(diffbuf, buflen + mb[i].size);
memcpy(diffbuf + buflen, mb[i].ptr, mb[i].size);
((linediff_fn)priv)(diffbuf, buflen + mb[i].size);
free(diffbuf);
diffbuf = NULL;
buflen = 0;
}
if (diffbuf) {
((linediff_fn)priv)(diffbuf, buflen);
free(diffbuf);
diffbuf = NULL;
buflen = 0;
}
return 0;
}
int cgit_diff_files(const struct object_id *old_oid,
const struct object_id *new_oid, unsigned long *old_size,
unsigned long *new_size, int *binary, int context,
int ignorews, linediff_fn fn)
{
mmfile_t file1, file2;
xpparam_t diff_params;
xdemitconf_t emit_params;
xdemitcb_t emit_cb;
if (!load_mmfile(&file1, old_oid) || !load_mmfile(&file2, new_oid))
return 1;
*old_size = file1.size;
*new_size = file2.size;
if ((file1.ptr && buffer_is_binary(file1.ptr, file1.size)) ||
(file2.ptr && buffer_is_binary(file2.ptr, file2.size))) {
*binary = 1;
if (file1.size)
free(file1.ptr);
if (file2.size)
free(file2.ptr);
return 0;
}
memset(&diff_params, 0, sizeof(diff_params));
memset(&emit_params, 0, sizeof(emit_params));
memset(&emit_cb, 0, sizeof(emit_cb));
diff_params.flags = XDF_NEED_MINIMAL;
if (ignorews)
diff_params.flags |= XDF_IGNORE_WHITESPACE;
emit_params.ctxlen = context > 0 ? context : 3;
emit_params.flags = XDL_EMIT_FUNCNAMES;
emit_cb.out_line = filediff_cb;
emit_cb.priv = fn;
xdl_diff(&file1, &file2, &diff_params, &emit_params, &emit_cb);
if (file1.size)
free(file1.ptr);
if (file2.size)
free(file2.ptr);
return 0;
}
void cgit_diff_tree(const struct object_id *old_oid,
const struct object_id *new_oid,
filepair_fn fn, const char *prefix, int ignorews)
{
struct diff_options opt;
struct pathspec_item *item;
repo_diff_setup(the_repository, &opt);
opt.output_format = DIFF_FORMAT_CALLBACK;
opt.detect_rename = 1;
opt.rename_limit = ctx.cfg.renamelimit;
opt.flags.recursive = 1;
if (ignorews)
DIFF_XDL_SET(&opt, IGNORE_WHITESPACE);
opt.format_callback = cgit_diff_tree_cb;
opt.format_callback_data = fn;
if (prefix) {
item = xcalloc(1, sizeof(*item));
item->match = xstrdup(prefix);
item->len = strlen(prefix);
opt.pathspec.nr = 1;
opt.pathspec.items = item;
}
diff_setup_done(&opt);
if (old_oid && !is_null_oid(old_oid))
diff_tree_oid(old_oid, new_oid, "", &opt);
else
diff_root_tree_oid(new_oid, "", &opt);
diffcore_std(&opt);
diff_flush(&opt);
}
void cgit_diff_commit(struct commit *commit, filepair_fn fn, const char *prefix)
{
const struct object_id *old_oid = NULL;
if (commit->parents)
old_oid = &commit->parents->item->object.oid;
cgit_diff_tree(old_oid, &commit->object.oid, fn, prefix,
ctx.qry.ignorews);
}
int cgit_parse_snapshots_mask(const char *str)
{
struct string_list tokens = STRING_LIST_INIT_DUP;
struct string_list_item *item;
const struct cgit_snapshot_format *f;
int rv = 0;
/* favor legacy setting */
if (atoi(str))
return 1;
if (strcmp(str, "all") == 0)
return INT_MAX;
string_list_split(&tokens, str, " ", -1);
string_list_remove_empty_items(&tokens, 0);
for_each_string_list_item(item, &tokens) {
for (f = cgit_snapshot_formats; f->suffix; f++) {
if (!strcmp(item->string, f->suffix) ||
!strcmp(item->string, f->suffix + 1)) {
rv |= cgit_snapshot_format_bit(f);
break;
}
}
}
string_list_clear(&tokens, 0);
return rv;
}
typedef struct {
char * name;
char * value;
} cgit_env_var;
void cgit_prepare_repo_env(struct cgit_repo * repo)
{
cgit_env_var env_vars[] = {
{ .name = "CGIT_REPO_URL", .value = repo->url },
{ .name = "CGIT_REPO_NAME", .value = repo->name },
{ .name = "CGIT_REPO_PATH", .value = repo->path },
{ .name = "CGIT_REPO_OWNER", .value = repo->owner },
{ .name = "CGIT_REPO_DEFBRANCH", .value = repo->defbranch },
{ .name = "CGIT_REPO_SECTION", .value = repo->section },
{ .name = "CGIT_REPO_CLONE_URL", .value = repo->clone_url }
};
int env_var_count = ARRAY_SIZE(env_vars);
cgit_env_var *p, *q;
static char *warn = "cgit warning: failed to set env: %s=%s\n";
p = env_vars;
q = p + env_var_count;
for (; p < q; p++)
if (p->value && setenv(p->name, p->value, 1))
fprintf(stderr, warn, p->name, p->value);
}
/* Read the content of the specified file into a newly allocated buffer,
* zeroterminate the buffer, truncate at a new line, and return 0 on success,
* errno otherwise.
*/
int read_first_line(const char *path, char **buf, size_t *size)
{
int fd, e;
struct stat st;
fd = open(path, O_RDONLY);
if (fd == -1)
return errno;
if (fstat(fd, &st)) {
e = errno;
close(fd);
return e;
}
if (!S_ISREG(st.st_mode)) {
close(fd);
return EISDIR;
}
*buf = xmalloc(st.st_size + 1);
*size = read_in_full(fd, *buf, st.st_size);
e = errno;
(*buf)[*size] = '\0';
*strchrnul(*buf, '\n') = '\0';
close(fd);
return (*size == st.st_size ? 0 : e);
}
char *strdup_first_line(const char *txt)
{
char *t = xstrdup(txt);
*strchrnul(t, '\n') = '\0';
return t;
}
static int is_token_char(char c)
{
return isalnum(c) || c == '_';
}
/* Replace name with getenv(name), return pointer to zero-terminating char
*/
static char *expand_macro(char *name, int maxlength)
{
char *value;
size_t len;
len = 0;
value = getenv(name);
if (value) {
len = strlen(value) + 1;
if (len > maxlength)
len = maxlength;
strlcpy(name, value, len);
--len;
}
return name + len;
}
#define EXPBUFSIZE (1024 * 8)
/* Replace all tokens prefixed by '$' in the specified text with the
* value of the named environment variable.
* NB: the return value is a static buffer, i.e. it must be strdup'd
* by the caller.
*/
char *expand_macros(const char *txt)
{
static char result[EXPBUFSIZE];
char *p, *start;
int len;
p = result;
start = NULL;
while (p < result + EXPBUFSIZE - 1 && txt && *txt) {
*p = *txt;
if (start) {
if (!is_token_char(*txt)) {
if (p - start > 0) {
*p = '\0';
len = result + EXPBUFSIZE - start - 1;
p = expand_macro(start, len) - 1;
}
start = NULL;
txt--;
}
p++;
txt++;
continue;
}
if (*txt == '$') {
start = p;
txt++;
continue;
}
p++;
txt++;
}
*p = '\0';
if (start && p - start > 0) {
len = result + EXPBUFSIZE - start - 1;
p = expand_macro(start, len);
*p = '\0';
}
return result;
}
char *get_mimetype_for_filename(const char *filename)
{
const char *ext;
char *mimetype, line[1024];
struct string_list list = STRING_LIST_INIT_NODUP;
int i;
FILE *file;
struct string_list_item *mime;
if (!filename)
return NULL;
ext = strrchr(filename, '.');
if (!ext)
return NULL;
++ext;
if (!ext[0])
return NULL;
mime = string_list_lookup(&ctx.cfg.mimetypes, ext);
if (mime)
return xstrdup(mime->util);
if (!ctx.cfg.mimetype_file)
return NULL;
file = fopen(ctx.cfg.mimetype_file, "r");
if (!file)
return NULL;
while (fgets(line, sizeof(line), file)) {
if (!line[0] || line[0] == '#')
continue;
string_list_split_in_place(&list, line, " \t\r\n", -1);
string_list_remove_empty_items(&list, 0);
mimetype = list.items[0].string;
for (i = 1; i < list.nr; i++) {
if (!strcasecmp(ext, list.items[i].string)) {
fclose(file);
return xstrdup(mimetype);
}
}
string_list_clear(&list, 0);
}
fclose(file);
return NULL;
}