cgit.c: add support for caching autodetected repositories

Signed-off-by: Lars Hjemli <hjemli@gmail.com>

2 files changed, 75 insertions, 1 deletions

diff --git a/cgit.c b/cgit.c
index db96905..594b950 100644
--- a/cgit.c
+++ b/cgit.c
@@ -11,162 +11,169 @@
 #include "cmd.h"
 #include "configfile.h"
 #include "html.h"
 #include "ui-shared.h"
 #include "ui-stats.h"
 #include "scan-tree.h"
 
 const char *cgit_version = CGIT_VERSION;
 
 void add_mimetype(const char *name, const char *value)
 {
         struct string_list_item *item;
 
         item = string_list_insert(xstrdup(name), &ctx.cfg.mimetypes);
         item->util = xstrdup(value);
 }
 
 struct cgit_filter *new_filter(const char *cmd, int extra_args)
 {
         struct cgit_filter *f;
 
         if (!cmd || !cmd[0])
                 return NULL;
 
         f = xmalloc(sizeof(struct cgit_filter));
         f->cmd = xstrdup(cmd);
         f->argv = xmalloc((2 + extra_args) * sizeof(char *));
         f->argv[0] = f->cmd;
         f->argv[1] = NULL;
         return f;
 }
 
+static void process_cached_repolist(const char *path);
+
 void config_cb(const char *name, const char *value)
 {
         if (!strcmp(name, "root-title"))
                 ctx.cfg.root_title = xstrdup(value);
         else if (!strcmp(name, "root-desc"))
                 ctx.cfg.root_desc = xstrdup(value);
         else if (!strcmp(name, "root-readme"))
                 ctx.cfg.root_readme = xstrdup(value);
         else if (!strcmp(name, "css"))
                 ctx.cfg.css = xstrdup(value);
         else if (!strcmp(name, "favicon"))
                 ctx.cfg.favicon = xstrdup(value);
         else if (!strcmp(name, "footer"))
                 ctx.cfg.footer = xstrdup(value);
         else if (!strcmp(name, "head-include"))
                 ctx.cfg.head_include = xstrdup(value);
         else if (!strcmp(name, "header"))
                 ctx.cfg.header = xstrdup(value);
         else if (!strcmp(name, "logo"))
                 ctx.cfg.logo = xstrdup(value);
         else if (!strcmp(name, "index-header"))
                 ctx.cfg.index_header = xstrdup(value);
         else if (!strcmp(name, "index-info"))
                 ctx.cfg.index_info = xstrdup(value);
         else if (!strcmp(name, "logo-link"))
                 ctx.cfg.logo_link = xstrdup(value);
         else if (!strcmp(name, "module-link"))
                 ctx.cfg.module_link = xstrdup(value);
         else if (!strcmp(name, "virtual-root")) {
                 ctx.cfg.virtual_root = trim_end(value, '/');
                 if (!ctx.cfg.virtual_root && (!strcmp(value, "/")))
                         ctx.cfg.virtual_root = "";
         } else if (!strcmp(name, "nocache"))
                 ctx.cfg.nocache = atoi(value);
         else if (!strcmp(name, "noplainemail"))
                 ctx.cfg.noplainemail = atoi(value);
         else if (!strcmp(name, "noheader"))
                 ctx.cfg.noheader = atoi(value);
         else if (!strcmp(name, "snapshots"))
                 ctx.cfg.snapshots = cgit_parse_snapshots_mask(value);
         else if (!strcmp(name, "enable-index-links"))
                 ctx.cfg.enable_index_links = atoi(value);
         else if (!strcmp(name, "enable-log-filecount"))
                 ctx.cfg.enable_log_filecount = atoi(value);
         else if (!strcmp(name, "enable-log-linecount"))
                 ctx.cfg.enable_log_linecount = atoi(value);
         else if (!strcmp(name, "max-stats"))
                 ctx.cfg.max_stats = cgit_find_stats_period(value, NULL);
         else if (!strcmp(name, "cache-size"))
                 ctx.cfg.cache_size = atoi(value);
         else if (!strcmp(name, "cache-root"))
                 ctx.cfg.cache_root = xstrdup(value);
         else if (!strcmp(name, "cache-root-ttl"))
                 ctx.cfg.cache_root_ttl = atoi(value);
         else if (!strcmp(name, "cache-repo-ttl"))
                 ctx.cfg.cache_repo_ttl = atoi(value);
+        else if (!strcmp(name, "cache-scanrc-ttl"))
+                ctx.cfg.cache_scanrc_ttl = atoi(value);
         else if (!strcmp(name, "cache-static-ttl"))
                 ctx.cfg.cache_static_ttl = atoi(value);
         else if (!strcmp(name, "cache-dynamic-ttl"))
                 ctx.cfg.cache_dynamic_ttl = atoi(value);
         else if (!strcmp(name, "about-filter"))
                 ctx.cfg.about_filter = new_filter(value, 0);
         else if (!strcmp(name, "commit-filter"))
                 ctx.cfg.commit_filter = new_filter(value, 0);
         else if (!strcmp(name, "embedded"))
                 ctx.cfg.embedded = atoi(value);
         else if (!strcmp(name, "max-message-length"))
                 ctx.cfg.max_msg_len = atoi(value);
         else if (!strcmp(name, "max-repodesc-length"))
                 ctx.cfg.max_repodesc_len = atoi(value);
         else if (!strcmp(name, "max-repo-count"))
                 ctx.cfg.max_repo_count = atoi(value);
         else if (!strcmp(name, "max-commit-count"))
                 ctx.cfg.max_commit_count = atoi(value);
         else if (!strcmp(name, "source-filter"))
                 ctx.cfg.source_filter = new_filter(value, 1);
         else if (!strcmp(name, "summary-log"))
                 ctx.cfg.summary_log = atoi(value);
         else if (!strcmp(name, "summary-branches"))
                 ctx.cfg.summary_branches = atoi(value);
         else if (!strcmp(name, "summary-tags"))
                 ctx.cfg.summary_tags = atoi(value);
         else if (!strcmp(name, "agefile"))
                 ctx.cfg.agefile = xstrdup(value);
         else if (!strcmp(name, "renamelimit"))
                 ctx.cfg.renamelimit = atoi(value);
         else if (!strcmp(name, "robots"))
                 ctx.cfg.robots = xstrdup(value);
         else if (!strcmp(name, "clone-prefix"))
                 ctx.cfg.clone_prefix = xstrdup(value);
         else if (!strcmp(name, "local-time"))
                 ctx.cfg.local_time = atoi(value);
         else if (!prefixcmp(name, "mimetype."))
                 add_mimetype(name + 9, value);
         else if (!strcmp(name, "repo.group"))
                 ctx.cfg.repo_group = xstrdup(value);
         else if (!strcmp(name, "repo.scan"))
-                scan_tree(value);
+                if (!ctx.cfg.nocache && ctx.cfg.cache_size)
+                        process_cached_repolist(value);
+                else
+                        scan_tree(value);
         else if (!strcmp(name, "repo.url"))
                 ctx.repo = cgit_add_repo(value);
         else if (!strcmp(name, "repo.name"))
                 ctx.repo->name = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.path"))
                 ctx.repo->path = trim_end(value, '/');
         else if (ctx.repo && !strcmp(name, "repo.clone-url"))
                 ctx.repo->clone_url = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.desc"))
                 ctx.repo->desc = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.owner"))
                 ctx.repo->owner = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.defbranch"))
                 ctx.repo->defbranch = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.snapshots"))
                 ctx.repo->snapshots = ctx.cfg.snapshots & cgit_parse_snapshots_mask(value); /* XXX: &? */
         else if (ctx.repo && !strcmp(name, "repo.enable-log-filecount"))
                 ctx.repo->enable_log_filecount = ctx.cfg.enable_log_filecount * atoi(value);
         else if (ctx.repo && !strcmp(name, "repo.enable-log-linecount"))
                 ctx.repo->enable_log_linecount = ctx.cfg.enable_log_linecount * atoi(value);
         else if (ctx.repo && !strcmp(name, "repo.max-stats"))
                 ctx.repo->max_stats = cgit_find_stats_period(value, NULL);
         else if (ctx.repo && !strcmp(name, "repo.module-link"))
                 ctx.repo->module_link= xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.about-filter"))
                 ctx.repo->about_filter = new_filter(value, 0);
         else if (ctx.repo && !strcmp(name, "repo.commit-filter"))
                 ctx.repo->commit_filter = new_filter(value, 0);
         else if (ctx.repo && !strcmp(name, "repo.source-filter"))
                 ctx.repo->source_filter = new_filter(value, 1);
         else if (ctx.repo && !strcmp(name, "repo.readme") && value != NULL) {
                 if (*value == '/')
@@ -207,64 +214,65 @@ static void querystring_cb(const char *name, const char *value)
                 ctx.qry.ofs = atoi(value);
         } else if (!strcmp(name, "path")) {
                 ctx.qry.path = trim_end(value, '/');
         } else if (!strcmp(name, "name")) {
                 ctx.qry.name = xstrdup(value);
         } else if (!strcmp(name, "mimetype")) {
                 ctx.qry.mimetype = xstrdup(value);
         } else if (!strcmp(name, "s")){
                 ctx.qry.sort = xstrdup(value);
         } else if (!strcmp(name, "showmsg")) {
                 ctx.qry.showmsg = atoi(value);
         } else if (!strcmp(name, "period")) {
                 ctx.qry.period = xstrdup(value);
         }
 }
 
 char *xstrdupn(const char *str)
 {
         return (str ? xstrdup(str) : NULL);
 }
 
 static void prepare_context(struct cgit_context *ctx)
 {
         memset(ctx, 0, sizeof(ctx));
         ctx->cfg.agefile = "info/web/last-modified";
         ctx->cfg.nocache = 0;
         ctx->cfg.cache_size = 0;
         ctx->cfg.cache_dynamic_ttl = 5;
         ctx->cfg.cache_max_create_time = 5;
         ctx->cfg.cache_repo_ttl = 5;
         ctx->cfg.cache_root = CGIT_CACHE_ROOT;
         ctx->cfg.cache_root_ttl = 5;
+        ctx->cfg.cache_scanrc_ttl = 15;
         ctx->cfg.cache_static_ttl = -1;
         ctx->cfg.css = "/cgit.css";
         ctx->cfg.logo = "/cgit.png";
         ctx->cfg.local_time = 0;
         ctx->cfg.max_repo_count = 50;
         ctx->cfg.max_commit_count = 50;
         ctx->cfg.max_lock_attempts = 5;
         ctx->cfg.max_msg_len = 80;
         ctx->cfg.max_repodesc_len = 80;
         ctx->cfg.max_stats = 0;
         ctx->cfg.module_link = "./?repo=%s&page=commit&id=%s";
         ctx->cfg.renamelimit = -1;
         ctx->cfg.robots = "index, nofollow";
         ctx->cfg.root_title = "Git repository browser";
         ctx->cfg.root_desc = "a fast webinterface for the git dscm";
         ctx->cfg.script_name = CGIT_SCRIPT_NAME;
         ctx->cfg.summary_branches = 10;
         ctx->cfg.summary_log = 10;
         ctx->cfg.summary_tags = 10;
         ctx->env.cgit_config = xstrdupn(getenv("CGIT_CONFIG"));
         ctx->env.http_host = xstrdupn(getenv("HTTP_HOST"));
         ctx->env.https = xstrdupn(getenv("HTTPS"));
         ctx->env.no_http = xstrdupn(getenv("NO_HTTP"));
         ctx->env.path_info = xstrdupn(getenv("PATH_INFO"));
         ctx->env.query_string = xstrdupn(getenv("QUERY_STRING"));
         ctx->env.request_method = xstrdupn(getenv("REQUEST_METHOD"));
         ctx->env.script_name = xstrdupn(getenv("SCRIPT_NAME"));
         ctx->env.server_name = xstrdupn(getenv("SERVER_NAME"));
         ctx->env.server_port = xstrdupn(getenv("SERVER_PORT"));
         ctx->page.mimetype = "text/html";
         ctx->page.charset = PAGE_ENCODING;
         ctx->page.filename = NULL;
@@ -409,64 +417,129 @@ static void process_request(void *cbdata)
         if (cmd->want_layout)
                 cgit_print_docend();
 }
 
 int cmp_repos(const void *a, const void *b)
 {
         const struct cgit_repo *ra = a, *rb = b;
         return strcmp(ra->url, rb->url);
 }
 
 void print_repo(FILE *f, struct cgit_repo *repo)
 {
         fprintf(f, "repo.url=%s\n", repo->url);
         fprintf(f, "repo.name=%s\n", repo->name);
         fprintf(f, "repo.path=%s\n", repo->path);
         if (repo->owner)
                 fprintf(f, "repo.owner=%s\n", repo->owner);
         if (repo->desc)
                 fprintf(f, "repo.desc=%s\n", repo->desc);
         if (repo->readme)
                 fprintf(f, "repo.readme=%s\n", repo->readme);
         fprintf(f, "\n");
 }
 
 void print_repolist(FILE *f, struct cgit_repolist *list, int start)
 {
         int i;
 
         for(i = start; i < list->count; i++)
                 print_repo(f, &list->repos[i]);
 }
 
+/* Scan 'path' for git repositories, save the resulting repolist in 'cached_rc'
+ * and return 0 on success.
+ */
+static int generate_cached_repolist(const char *path, const char *cached_rc)
+{
+        char *locked_rc;
+        int idx;
+        FILE *f;
+
+        locked_rc = xstrdup(fmt("%s.lock", cached_rc));
+        f = fopen(locked_rc, "wx");
+        if (!f) {
+                /* Inform about the error unless the lockfile already existed,
+                 * since that only means we've got concurrent requests.
+                 */
+                if (errno != EEXIST)
+                        fprintf(stderr, "[cgit] Error opening %s: %s (%d)\n",
+                                locked_rc, strerror(errno), errno);
+                return errno;
+        }
+        idx = cgit_repolist.count;
+        scan_tree(path);
+        print_repolist(f, &cgit_repolist, idx);
+        if (rename(locked_rc, cached_rc))
+                fprintf(stderr, "[cgit] Error renaming %s to %s: %s (%d)\n",
+                        locked_rc, cached_rc, strerror(errno), errno);
+        fclose(f);
+        return 0;
+}
+
+static void process_cached_repolist(const char *path)
+{
+        struct stat st;
+        char *cached_rc;
+        time_t age;
+
+        cached_rc = xstrdup(fmt("%s/rc-%8x", ctx.cfg.cache_root,
+                hash_str(path)));
+
+        if (stat(cached_rc, &st)) {
+                /* Nothing is cached, we need to scan without forking. And
+                 * if we fail to generate a cached repolist, we need to
+                 * invoke scan_tree manually.
+                 */
+                if (generate_cached_repolist(path, cached_rc))
+                        scan_tree(path);
+                return;
+        }
+
+        parse_configfile(cached_rc, config_cb);
+
+        /* If the cached configfile hasn't expired, lets exit now */
+        age = time(NULL) - st.st_mtime;
+        if (age <= (ctx.cfg.cache_scanrc_ttl * 60))
+                return;
+
+        /* The cached repolist has been parsed, but it was old. So lets
+         * rescan the specified path and generate a new cached repolist
+         * in a child-process to avoid latency for the current request.
+         */
+        if (fork())
+                return;
+
+        exit(generate_cached_repolist(path, cached_rc));
+}
 
 static void cgit_parse_args(int argc, const char **argv)
 {
         int i;
         int scan = 0;
 
         for (i = 1; i < argc; i++) {
                 if (!strncmp(argv[i], "--cache=", 8)) {
                         ctx.cfg.cache_root = xstrdup(argv[i]+8);
                 }
                 if (!strcmp(argv[i], "--nocache")) {
                         ctx.cfg.nocache = 1;
                 }
                 if (!strcmp(argv[i], "--nohttp")) {
                         ctx.env.no_http = "1";
                 }
                 if (!strncmp(argv[i], "--query=", 8)) {
                         ctx.qry.raw = xstrdup(argv[i]+8);
                 }
                 if (!strncmp(argv[i], "--repo=", 7)) {
                         ctx.qry.repo = xstrdup(argv[i]+7);
                 }
                 if (!strncmp(argv[i], "--page=", 7)) {
                         ctx.qry.page = xstrdup(argv[i]+7);
                 }
                 if (!strncmp(argv[i], "--head=", 7)) {
                         ctx.qry.head = xstrdup(argv[i]+7);
                         ctx.qry.has_symref = 1;
                 }
                 if (!strncmp(argv[i], "--sha1=", 7)) {
                         ctx.qry.sha1 = xstrdup(argv[i]+7);
                         ctx.qry.has_sha1 = 1;
diff --git a/cgit.h b/cgit.h
index adb8da4..5659580 100644
--- a/cgit.h
+++ b/cgit.h
@@ -139,64 +139,65 @@ struct cgit_query {
         int   ofs;
         int nohead;
         char *sort;
         int showmsg;
 };
 
 struct cgit_config {
         char *agefile;
         char *cache_root;
         char *clone_prefix;
         char *css;
         char *favicon;
         char *footer;
         char *head_include;
         char *header;
         char *index_header;
         char *index_info;
         char *logo;
         char *logo_link;
         char *module_link;
         char *repo_group;
         char *robots;
         char *root_title;
         char *root_desc;
         char *root_readme;
         char *script_name;
         char *virtual_root;
         int cache_size;
         int cache_dynamic_ttl;
         int cache_max_create_time;
         int cache_repo_ttl;
         int cache_root_ttl;
+        int cache_scanrc_ttl;
         int cache_static_ttl;
         int embedded;
         int enable_index_links;
         int enable_log_filecount;
         int enable_log_linecount;
         int local_time;
         int max_repo_count;
         int max_commit_count;
         int max_lock_attempts;
         int max_msg_len;
         int max_repodesc_len;
         int max_stats;
         int nocache;
         int noplainemail;
         int noheader;
         int renamelimit;
         int snapshots;
         int summary_branches;
         int summary_log;
         int summary_tags;
         struct string_list mimetypes;
         struct cgit_filter *about_filter;
         struct cgit_filter *commit_filter;
         struct cgit_filter *source_filter;
 };
 
 struct cgit_page {
         time_t modified;
         time_t expires;
         size_t size;
         char *mimetype;
         char *charset;