Add caching infrastructure

This enables internal caching of page output.

Page requests are split into four groups:
  1) repo listing (front page)
  2) repo summary
  3) repo pages w/symbolic references in query string
  4) repo pages w/constant sha1's in query string

Each group has a TTL specified in minutes. When a page is requested, a cached
filename is stat(2)'ed and st_mtime is compared to time(2). If TTL has expired
(or the file didn't exist), the cached file is regenerated.

When generating a cached file, locking is used to avoid parallell processing
of the request. If multiple processes tries to aquire the same lock, the ones
who fail to get the lock serves the (expired) cached file. If the cached file
don't exist, the process instead calls sched_yield(2) before restarting the
request processing.

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

1 files changed, 102 insertions, 15 deletions

diff --git a/cgit.c b/cgit.c
index 4c14f77..09c857c 100644
--- a/cgit.c
+++ b/cgit.c
@@ -12,4 +12,5 @@ static const char cgit_lib_error[] =
 
+int htmlfd = 0;
 
-char *cgit_root         = "/var/git";
+char *cgit_root         = "/usr/src/git";
 char *cgit_root_title   = "Git repository browser";
@@ -20,2 +21,10 @@ char *cgit_virtual_root = NULL;
 
+char *cgit_cache_root   = "/var/cache/cgit";
+
+int cgit_cache_root_ttl        =  5;
+int cgit_cache_repo_ttl        =  5;
+int cgit_cache_dynamic_ttl     =  5;
+int cgit_cache_static_ttl      = -1;
+int cgit_cache_max_create_time =  5;
+
 char *cgit_repo_name    = NULL;
@@ -24,2 +33,6 @@ char *cgit_repo_owner   = NULL;
 
+int cgit_query_has_symref = 0;
+int cgit_query_has_sha1   = 0;
+
+char *cgit_querystring  = NULL;
 char *cgit_query_repo   = NULL;
@@ -27,2 +40,5 @@ char *cgit_query_page   = NULL;
 char *cgit_query_head   = NULL;
+char *cgit_query_sha1   = NULL;
+
+struct cacheitem cacheitem;
 
@@ -30,3 +46,3 @@ int cgit_parse_query(char *txt, configfn fn)
 {
-        char *t = txt, *value = NULL, c;
+        char *t, *value = NULL, c;
 
@@ -35,2 +51,4 @@ int cgit_parse_query(char *txt, configfn fn)
 
+        t = txt = xstrdup(txt);
+ 
         while((c=*t) != '\0') {
@@ -84,4 +102,9 @@ void cgit_querystring_cb(const char *name, const char *value)
                 cgit_query_page = xstrdup(value);
-        else if (!strcmp(name, "h"))
+        else if (!strcmp(name, "h")) {
                 cgit_query_head = xstrdup(value);
+                cgit_query_has_symref = 1;
+        } else if (!strcmp(name, "id")) {
+                cgit_query_sha1 = xstrdup(value);
+                cgit_query_has_sha1 = 1;
+        }
 }
@@ -138,2 +161,22 @@ static int cgit_print_branch_cb(const char *refname, const unsigned char *sha1,
 
+/* Sun, 06 Nov 1994 08:49:37 GMT */
+static char *http_date(time_t t)
+{
+        static char day[][4] = {"Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Sat"};
+        static char month[][4] = {"Jan", "Feb", "Mar", "Apr", "May", "Jun",
+                                  "Jul", "Aug", "Sep", "Oct", "Now", "Dec"};
+        struct tm *tm = gmtime(&t);
+        return fmt("%s, %02d %s %04d %02d:%02d:%02d GMT", day[tm->tm_wday],
+                   tm->tm_mday, month[tm->tm_mon], 1900+tm->tm_year,
+                   tm->tm_hour, tm->tm_min, tm->tm_sec);
+}
+
+static int ttl_seconds(int ttl)
+{
+        if (ttl<0)
+                return 60 * 60 * 24 * 365;
+        else 
+                return ttl * 60;
+}
+
 static void cgit_print_docstart(char *title)
@@ -141,2 +184,5 @@ static void cgit_print_docstart(char *title)
         html("Content-Type: text/html; charset=utf-8\n");
+        htmlf("Last-Modified: %s\n", http_date(cacheitem.st.st_mtime));
+        htmlf("Expires: %s\n", http_date(cacheitem.st.st_mtime + 
+                                         ttl_seconds(cacheitem.ttl)));
         html("\n");
@@ -177,2 +223,3 @@ static void cgit_print_repolist()
 
+        chdir(cgit_root);
         cgit_print_docstart(cgit_root_title);
@@ -199,3 +246,3 @@ static void cgit_print_repolist()
                 cgit_repo_name = cgit_repo_desc = cgit_repo_owner = NULL;
-                name = fmt("%s/.git/info/cgit", de->d_name);
+                name = fmt("%s/info/cgit", de->d_name);
                 if (cgit_read_config(name, cgit_repo_config_cb))
@@ -293,3 +340,3 @@ static void cgit_print_commit_shortlog(struct commit *commit)
         html("</td><td>");
-        char *qry = fmt("h=%s", sha1_to_hex(commit->object.sha1));
+        char *qry = fmt("id=%s", sha1_to_hex(commit->object.sha1));
         char *url = cgit_pageurl(cgit_query_repo, "view", qry);
@@ -373,4 +420,4 @@ static void cgit_print_repo_page()
 {
-        if (chdir(cgit_query_repo) || 
-            cgit_read_config(".git/info/cgit", cgit_repo_config_cb)) {
+        if (chdir(fmt("%s/%s", cgit_root, cgit_query_repo)) || 
+            cgit_read_config("info/cgit", cgit_repo_config_cb)) {
                 char *title = fmt("%s - %s", cgit_root_title, "Bad request");
@@ -383,3 +430,3 @@ static void cgit_print_repo_page()
         }
-        
+        setenv("GIT_DIR", fmt("%s/%s", cgit_root, cgit_query_repo), 1);
         char *title = fmt("%s - %s", cgit_repo_name, cgit_repo_desc);
@@ -392,3 +439,3 @@ static void cgit_print_repo_page()
         } else if (!strcmp(cgit_query_page, "view")) {
-                cgit_print_object(cgit_query_head);
+                cgit_print_object(cgit_query_sha1);
         }
@@ -397,9 +444,6 @@ static void cgit_print_repo_page()
 
-int main(int argc, const char **argv)
+static void cgit_fill_cache(struct cacheitem *item)
 {
-        if (cgit_read_config("/etc/cgitrc", cgit_global_config_cb))
-                die("Error reading config: %d %s", errno, strerror(errno));
-
-        chdir(cgit_root);
-        cgit_parse_query(getenv("QUERY_STRING"), cgit_querystring_cb);
+        htmlfd = item->fd;
+        item->st.st_mtime = time(NULL);
         if (cgit_query_repo)
@@ -408,2 +452,45 @@ int main(int argc, const char **argv)
                 cgit_print_repolist();
+}
+
+static void cgit_refresh_cache(struct cacheitem *item)
+{
+ top:
+        if (!cache_lookup(item)) {
+                if (cache_lock(item)) {
+                        cgit_fill_cache(item);
+                        cache_unlock(item);
+                } else {
+                        sched_yield();
+                        goto top;
+                }
+        } else if (cache_expired(item)) {
+                if (cache_lock(item)) {
+                        cgit_fill_cache(item);
+                        cache_unlock(item);
+                }
+        }
+}
+
+static void cgit_print_cache(struct cacheitem *item)
+{
+        static char buf[4096];
+        ssize_t i;
+
+        int fd = open(item->name, O_RDONLY);
+        if (fd<0)
+                die("Unable to open cached file %s", item->name);
+
+        while((i=read(fd, buf, sizeof(buf))) > 0)
+                write(STDOUT_FILENO, buf, i);
+
+        close(fd);
+}
+
+int main(int argc, const char **argv)
+{
+        cgit_read_config("/etc/cgitrc", cgit_global_config_cb);
+        cgit_querystring = xstrdup(getenv("QUERY_STRING"));
+        cgit_parse_query(cgit_querystring, cgit_querystring_cb);
+        cgit_refresh_cache(&cacheitem);
+        cgit_print_cache(&cacheitem);
         return 0;