Add support for ETag in 'plain' view

When downloading a blob identified by its path, the client might want
to know if the blob has been modified since a previous download of the
same path. To this end, an ETag containing the blob SHA1 seems to be
ideal.

Todo: add support for HEAD requests...

Suggested-by: Owen Taylor <otaylor@redhat.com>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>

4 files changed, 5 insertions, 0 deletions

diff --git a/cgit.c b/cgit.c
index 64d95f9..411e421 100644
--- a/cgit.c
+++ b/cgit.c
@@ -113,192 +113,193 @@ void config_cb(const char *name, const char *value)
                 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.readme") && value != NULL) {
                 if (*value == '/')
                         ctx.repo->readme = xstrdup(value);
                 else
                         ctx.repo->readme = xstrdup(fmt("%s/%s", ctx.repo->path, value));
         } else if (!strcmp(name, "include"))
                 parse_configfile(value, config_cb);
 }
 
 static void querystring_cb(const char *name, const char *value)
 {
         if (!strcmp(name,"r")) {
                 ctx.qry.repo = xstrdup(value);
                 ctx.repo = cgit_get_repoinfo(value);
         } else if (!strcmp(name, "p")) {
                 ctx.qry.page = xstrdup(value);
         } else if (!strcmp(name, "url")) {
                 ctx.qry.url = xstrdup(value);
                 cgit_parse_url(value);
         } else if (!strcmp(name, "qt")) {
                 ctx.qry.grep = xstrdup(value);
         } else if (!strcmp(name, "q")) {
                 ctx.qry.search = xstrdup(value);
         } else if (!strcmp(name, "h")) {
                 ctx.qry.head = xstrdup(value);
                 ctx.qry.has_symref = 1;
         } else if (!strcmp(name, "id")) {
                 ctx.qry.sha1 = xstrdup(value);
                 ctx.qry.has_sha1 = 1;
         } else if (!strcmp(name, "id2")) {
                 ctx.qry.sha2 = xstrdup(value);
                 ctx.qry.has_sha1 = 1;
         } else if (!strcmp(name, "ofs")) {
                 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);
         }
 }
 
 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_static_ttl = -1;
         ctx->cfg.css = "/cgit.css";
         ctx->cfg.logo = "/git-logo.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->page.mimetype = "text/html";
         ctx->page.charset = PAGE_ENCODING;
         ctx->page.filename = NULL;
         ctx->page.size = 0;
         ctx->page.modified = time(NULL);
         ctx->page.expires = ctx->page.modified;
+        ctx->page.etag = NULL;
 }
 
 struct refmatch {
         char *req_ref;
         char *first_ref;
         int match;
 };
 
 int find_current_ref(const char *refname, const unsigned char *sha1,
                      int flags, void *cb_data)
 {
         struct refmatch *info;
 
         info = (struct refmatch *)cb_data;
         if (!strcmp(refname, info->req_ref))
                 info->match = 1;
         if (!info->first_ref)
                 info->first_ref = xstrdup(refname);
         return info->match;
 }
 
 char *find_default_branch(struct cgit_repo *repo)
 {
         struct refmatch info;
         char *ref;
 
         info.req_ref = repo->defbranch;
         info.first_ref = NULL;
         info.match = 0;
         for_each_branch_ref(find_current_ref, &info);
         if (info.match)
                 ref = info.req_ref;
         else
                 ref = info.first_ref;
         if (ref)
                 ref = xstrdup(ref);
         return ref;
 }
 
 static int prepare_repo_cmd(struct cgit_context *ctx)
 {
         char *tmp;
         unsigned char sha1[20];
         int nongit = 0;
 
         setenv("GIT_DIR", ctx->repo->path, 1);
         setup_git_directory_gently(&nongit);
         if (nongit) {
                 ctx->page.title = fmt("%s - %s", ctx->cfg.root_title,
                                       "config error");
                 tmp = fmt("Not a git repository: '%s'", ctx->repo->path);
                 ctx->repo = NULL;
                 cgit_print_http_headers(ctx);
                 cgit_print_docstart(ctx);
                 cgit_print_pageheader(ctx);
                 cgit_print_error(tmp);
                 cgit_print_docend();
                 return 1;
         }
         ctx->page.title = fmt("%s - %s", ctx->repo->name, ctx->repo->desc);
 
         if (!ctx->qry.head) {
                 ctx->qry.nohead = 1;
                 ctx->qry.head = find_default_branch(ctx->repo);
                 ctx->repo->defbranch = ctx->qry.head;
         }
 
         if (!ctx->qry.head) {
                 cgit_print_http_headers(ctx);
                 cgit_print_docstart(ctx);
                 cgit_print_pageheader(ctx);
                 cgit_print_error("Repository seems to be empty");
                 cgit_print_docend();
                 return 1;
         }
 
         if (get_sha1(ctx->qry.head, sha1)) {
                 tmp = xstrdup(ctx->qry.head);
                 ctx->qry.head = ctx->repo->defbranch;
                 cgit_print_http_headers(ctx);
                 cgit_print_docstart(ctx);
                 cgit_print_pageheader(ctx);
                 cgit_print_error(fmt("Invalid branch: %s", tmp));
                 cgit_print_docend();
                 return 1;
         }
         return 0;
 }
 
 static void process_request(void *cbdata)
 {
         struct cgit_context *ctx = cbdata;
         struct cgit_cmd *cmd;
 
         cmd = cgit_get_cmd(ctx);
         if (!cmd) {
diff --git a/cgit.h b/cgit.h
index 5f7af51..2bfbe7b 100644
--- a/cgit.h
+++ b/cgit.h
@@ -87,160 +87,161 @@ struct commitinfo {
 
 struct taginfo {
         char *tagger;
         char *tagger_email;
         unsigned long tagger_date;
         char *msg;
 };
 
 struct refinfo {
         const char *refname;
         struct object *object;
         union {
                 struct taginfo *tag;
                 struct commitinfo *commit;
         };
 };
 
 struct reflist {
         struct refinfo **refs;
         int alloc;
         int count;
 };
 
 struct cgit_query {
         int has_symref;
         int has_sha1;
         char *raw;
         char *repo;
         char *page;
         char *search;
         char *grep;
         char *head;
         char *sha1;
         char *sha2;
         char *path;
         char *name;
         char *mimetype;
         char *url;
         char *period;
         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 *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_static_ttl;
         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 renamelimit;
         int snapshots;
         int summary_branches;
         int summary_log;
         int summary_tags;
 };
 
 struct cgit_page {
         time_t modified;
         time_t expires;
         size_t size;
         char *mimetype;
         char *charset;
         char *filename;
+        char *etag;
         char *title;
 };
 
 struct cgit_context {
         struct cgit_query qry;
         struct cgit_config cfg;
         struct cgit_repo *repo;
         struct cgit_page page;
 };
 
 struct cgit_snapshot_format {
         const char *suffix;
         const char *mimetype;
         write_archive_fn_t write_func;
         int bit;
 };
 
 extern const char *cgit_version;
 
 extern struct cgit_repolist cgit_repolist;
 extern struct cgit_context ctx;
 extern const struct cgit_snapshot_format cgit_snapshot_formats[];
 
 extern struct cgit_repo *cgit_add_repo(const char *url);
 extern struct cgit_repo *cgit_get_repoinfo(const char *url);
 extern void cgit_repo_config_cb(const char *name, const char *value);
 
 extern int chk_zero(int result, char *msg);
 extern int chk_positive(int result, char *msg);
 extern int chk_non_negative(int result, char *msg);
 
 extern char *trim_end(const char *str, char c);
 extern char *strlpart(char *txt, int maxlen);
 extern char *strrpart(char *txt, int maxlen);
 
 extern void cgit_add_ref(struct reflist *list, struct refinfo *ref);
 extern int cgit_refs_cb(const char *refname, const unsigned char *sha1,
                         int flags, void *cb_data);
 
 extern void *cgit_free_commitinfo(struct commitinfo *info);
 
 extern int cgit_diff_files(const unsigned char *old_sha1,
                            const unsigned char *new_sha1,
                            unsigned long *old_size, unsigned long *new_size,
                            int *binary, linediff_fn fn);
 
 extern void cgit_diff_tree(const unsigned char *old_sha1,
                            const unsigned char *new_sha1,
                            filepair_fn fn, const char *prefix);
 
 extern void cgit_diff_commit(struct commit *commit, filepair_fn fn);
 
 extern char *fmt(const char *format,...);
 
 extern struct commitinfo *cgit_parse_commit(struct commit *commit);
 extern struct taginfo *cgit_parse_tag(struct tag *tag);
 extern void cgit_parse_url(const char *url);
 
 extern const char *cgit_repobasename(const char *reponame);
 
 extern int cgit_parse_snapshots_mask(const char *str);
 
 
 #endif /* CGIT_H */
diff --git a/ui-plain.c b/ui-plain.c
index 5addd9e..f73cd14 100644
--- a/ui-plain.c
+++ b/ui-plain.c
@@ -1,79 +1,80 @@
 /* ui-plain.c: functions for output of plain blobs by path
  *
  * Copyright (C) 2008 Lars Hjemli
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
  */
 
 #include "cgit.h"
 #include "html.h"
 #include "ui-shared.h"
 
 char *curr_rev;
 char *match_path;
 int match;
 
 static void print_object(const unsigned char *sha1, const char *path)
 {
         enum object_type type;
         char *buf;
         unsigned long size;
 
         type = sha1_object_info(sha1, &size);
         if (type == OBJ_BAD) {
                 html_status(404, "Not found", 0);
                 return;
         }
 
         buf = read_sha1_file(sha1, &type, &size);
         if (!buf) {
                 html_status(404, "Not found", 0);
                 return;
         }
         ctx.page.mimetype = "text/plain";
         ctx.page.filename = fmt("%s", path);
         ctx.page.size = size;
+        ctx.page.etag = sha1_to_hex(sha1);
         cgit_print_http_headers(&ctx);
         html_raw(buf, size);
         match = 1;
 }
 
 static int walk_tree(const unsigned char *sha1, const char *base, int baselen,
                      const char *pathname, unsigned mode, int stage,
                      void *cbdata)
 {
         if (S_ISDIR(mode))
                 return READ_TREE_RECURSIVE;
 
         if (S_ISREG(mode))
                 print_object(sha1, pathname);
 
         return 0;
 }
 
 void cgit_print_plain(struct cgit_context *ctx)
 {
         const char *rev = ctx->qry.sha1;
         unsigned char sha1[20];
         struct commit *commit;
         const char *paths[] = {ctx->qry.path, NULL};
 
         if (!rev)
                 rev = ctx->qry.head;
 
         curr_rev = xstrdup(rev);
         if (get_sha1(rev, sha1)) {
                 html_status(404, "Not found", 0);
                 return;
         }
         commit = lookup_commit_reference(sha1);
         if (!commit || parse_commit(commit)) {
                 html_status(404, "Not found", 0);
                 return;
         }
         match_path = ctx->qry.path;
         read_tree_recursive(commit->tree, NULL, 0, 0, paths, walk_tree, NULL);
         if (!match)
                 html_status(404, "Not found", 0);
 }
diff --git a/ui-shared.c b/ui-shared.c
index de77bbf..86a7d29 100644
--- a/ui-shared.c
+++ b/ui-shared.c
@@ -375,192 +375,194 @@ void cgit_stats_link(char *name, char *title, char *class, char *head,
         reporevlink("stats", name, title, class, head, NULL, path);
 }
 
 void cgit_object_link(struct object *obj)
 {
         char *page, *shortrev, *fullrev, *name;
 
         fullrev = sha1_to_hex(obj->sha1);
         shortrev = xstrdup(fullrev);
         shortrev[10] = '\0';
         if (obj->type == OBJ_COMMIT) {
                 cgit_commit_link(fmt("commit %s...", shortrev), NULL, NULL,
                                  ctx.qry.head, fullrev);
                 return;
         } else if (obj->type == OBJ_TREE)
                 page = "tree";
         else if (obj->type == OBJ_TAG)
                 page = "tag";
         else
                 page = "blob";
         name = fmt("%s %s...", typename(obj->type), shortrev);
         reporevlink(page, name, NULL, NULL, ctx.qry.head, fullrev, NULL);
 }
 
 void cgit_print_date(time_t secs, char *format, int local_time)
 {
         char buf[64];
         struct tm *time;
 
         if (!secs)
                 return;
         if(local_time)
                 time = localtime(&secs);
         else
                 time = gmtime(&secs);
         strftime(buf, sizeof(buf)-1, format, time);
         html_txt(buf);
 }
 
 void cgit_print_age(time_t t, time_t max_relative, char *format)
 {
         time_t now, secs;
 
         if (!t)
                 return;
         time(&now);
         secs = now - t;
 
         if (secs > max_relative && max_relative >= 0) {
                 cgit_print_date(t, format, ctx.cfg.local_time);
                 return;
         }
 
         if (secs < TM_HOUR * 2) {
                 htmlf("<span class='age-mins'>%.0f min.</span>",
                       secs * 1.0 / TM_MIN);
                 return;
         }
         if (secs < TM_DAY * 2) {
                 htmlf("<span class='age-hours'>%.0f hours</span>",
                       secs * 1.0 / TM_HOUR);
                 return;
         }
         if (secs < TM_WEEK * 2) {
                 htmlf("<span class='age-days'>%.0f days</span>",
                       secs * 1.0 / TM_DAY);
                 return;
         }
         if (secs < TM_MONTH * 2) {
                 htmlf("<span class='age-weeks'>%.0f weeks</span>",
                       secs * 1.0 / TM_WEEK);
                 return;
         }
         if (secs < TM_YEAR * 2) {
                 htmlf("<span class='age-months'>%.0f months</span>",
                       secs * 1.0 / TM_MONTH);
                 return;
         }
         htmlf("<span class='age-years'>%.0f years</span>",
               secs * 1.0 / TM_YEAR);
 }
 
 void cgit_print_http_headers(struct cgit_context *ctx)
 {
         if (ctx->page.mimetype && ctx->page.charset)
                 htmlf("Content-Type: %s; charset=%s\n", ctx->page.mimetype,
                       ctx->page.charset);
         else if (ctx->page.mimetype)
                 htmlf("Content-Type: %s\n", ctx->page.mimetype);
         if (ctx->page.size)
                 htmlf("Content-Length: %ld\n", ctx->page.size);
         if (ctx->page.filename)
                 htmlf("Content-Disposition: inline; filename=\"%s\"\n",
                       ctx->page.filename);
         htmlf("Last-Modified: %s\n", http_date(ctx->page.modified));
         htmlf("Expires: %s\n", http_date(ctx->page.expires));
+        if (ctx->page.etag)
+                htmlf("ETag: \"%s\"\n", ctx->page.etag);
         html("\n");
 }
 
 void cgit_print_docstart(struct cgit_context *ctx)
 {
         char *host = cgit_hosturl();
         html(cgit_doctype);
         html("<html xmlns='http://www.w3.org/1999/xhtml' xml:lang='en' lang='en'>\n");
         html("<head>\n");
         html("<title>");
         html_txt(ctx->page.title);
         html("</title>\n");
         htmlf("<meta name='generator' content='cgit %s'/>\n", cgit_version);
         if (ctx->cfg.robots && *ctx->cfg.robots)
                 htmlf("<meta name='robots' content='%s'/>\n", ctx->cfg.robots);
         html("<link rel='stylesheet' type='text/css' href='");
         html_attr(ctx->cfg.css);
         html("'/>\n");
         if (ctx->cfg.favicon) {
                 html("<link rel='shortcut icon' href='");
                 html_attr(ctx->cfg.favicon);
                 html("'/>\n");
         }
         if (host && ctx->repo) {
                 html("<link rel='alternate' title='Atom feed' href='http://");
                 html_attr(cgit_hosturl());
                 html_attr(cgit_fileurl(ctx->repo->url, "atom", ctx->qry.path,
                                        fmt("h=%s", ctx->qry.head)));
                 html("' type='application/atom+xml'/>");
         }
         html("</head>\n");
         html("<body>\n");
         if (ctx->cfg.header)
                 html_include(ctx->cfg.header);
 }
 
 void cgit_print_docend()
 {
         html("</div>");
         if (ctx.cfg.footer)
                 html_include(ctx.cfg.footer);
         else {
                 htmlf("<div class='footer'>generated  by cgit %s at ",
                         cgit_version);
                 cgit_print_date(time(NULL), FMT_LONGDATE, ctx.cfg.local_time);
                 html("</div>\n");
         }
         html("</body>\n</html>\n");
 }
 
 int print_branch_option(const char *refname, const unsigned char *sha1,
                         int flags, void *cb_data)
 {
         char *name = (char *)refname;
         html_option(name, name, ctx.qry.head);
         return 0;
 }
 
 int print_archive_ref(const char *refname, const unsigned char *sha1,
                        int flags, void *cb_data)
 {
         struct tag *tag;
         struct taginfo *info;
         struct object *obj;
         char buf[256], *url;
         unsigned char fileid[20];
         int *header = (int *)cb_data;
 
         if (prefixcmp(refname, "refs/archives"))
                 return 0;
         strncpy(buf, refname+14, sizeof(buf));
         obj = parse_object(sha1);
         if (!obj)
                 return 1;
         if (obj->type == OBJ_TAG) {
                 tag = lookup_tag(sha1);
                 if (!tag || parse_tag(tag) || !(info = cgit_parse_tag(tag)))
                         return 0;
                 hashcpy(fileid, tag->tagged->sha1);
         } else if (obj->type != OBJ_BLOB) {
                 return 0;
         } else {
                 hashcpy(fileid, sha1);
         }
         if (!*header) {
                 html("<h1>download</h1>\n");
                 *header = 1;
         }
         url = cgit_pageurl(ctx.qry.repo, "blob",
                            fmt("id=%s&amp;path=%s", sha1_to_hex(fileid),
                                buf));
         html_link_open(url, NULL, "menu");
         html_txt(strlpart(buf, 20));
         html_link_close();
         return 0;
 }