readme: use string_list instead of space deliminations

Now this is possible in cgitrc -

readme=:README.md
readme=:readme.md
readme=:README.mkd
readme=:readme.mkd
readme=:README.rst
readme=:readme.rst
readme=:README.html
readme=:readme.html
readme=:README.htm
readme=:readme.htm
readme=:README.txt
readme=:readme.txt
readme=:README
readme=:readme
readme=:INSTALL.txt
readme=:install.txt
readme=:INSTALL
readme=:install

Suggested-by: John Keeping <john@keeping.me.uk>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>

10 files changed, 160 insertions, 122 deletions

diff --git a/cgit.c b/cgit.c
index f738b83..f0a9acf 100644
--- a/cgit.c
+++ b/cgit.c
@@ -1,7 +1,7 @@
 /* cgit.c: cgi for the git scm
  *
  * Copyright (C) 2006 Lars Hjemli
- * Copyright (C) 2010, 2012 Jason A. Donenfeld <Jason@zx2c4.com>
+ * Copyright (C) 2010-2013 Jason A. Donenfeld <Jason@zx2c4.com>
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
@@ -101,13 +101,15 @@ static void repo_config(struct cgit_repo *repo, const char *name, const char *va
         else if (!strcmp(name, "module-link"))
                 repo->module_link= xstrdup(value);
         else if (!prefixcmp(name, "module-link.")) {
-                item = string_list_append(&repo->submodules, name + 12);
+                item = string_list_append(&repo->submodules, xstrdup(name + 12));
                 item->util = xstrdup(value);
         } else if (!strcmp(name, "section"))
                 repo->section = xstrdup(value);
-        else if (!strcmp(name, "readme") && value != NULL)
-                repo->readme = xstrdup(value);
-        else if (!strcmp(name, "logo") && value != NULL)
+        else if (!strcmp(name, "readme") && value != NULL) {
+                if (repo->readme.items == ctx.cfg.readme.items)
+                        memset(&repo->readme, 0, sizeof(repo->readme));
+                string_list_append(&repo->readme, xstrdup(value));
+        } else if (!strcmp(name, "logo") && value != NULL)
                 repo->logo = xstrdup(value);
         else if (!strcmp(name, "logo-link") && value != NULL)
                 repo->logo_link = xstrdup(value);
@@ -131,8 +133,8 @@ static void config_cb(const char *name, const char *value)
                 ctx.repo->path = trim_end(value, '/');
         else if (ctx.repo && !prefixcmp(name, "repo."))
                 repo_config(ctx.repo, name + 5, value);
-        else if (!strcmp(name, "readme"))
-                ctx.cfg.readme = xstrdup(value);
+        else if (!strcmp(name, "readme") && value != NULL)
+                string_list_append(&ctx.cfg.readme, xstrdup(value));
         else if (!strcmp(name, "root-title"))
                 ctx.cfg.root_title = xstrdup(value);
         else if (!strcmp(name, "root-desc"))
@@ -470,37 +472,76 @@ static char *guess_defbranch(void)
                 return "master";
         return xstrdup(ref + 11);
 }
+/* The caller must free filename and ref after calling this. */
+static inline void parse_readme(const char *readme, char **filename, char **ref, struct cgit_repo *repo)
+{
+        const char *colon;
+
+        *filename = NULL;
+        *ref = NULL;
+
+        if (!readme || !readme[0])
+                return;
 
+        /* Check if the readme is tracked in the git repo. */
+        colon = strchr(readme, ':');
+        if (colon && strlen(colon) > 1) {
+                /* If it starts with a colon, we want to use
+                 * the default branch */
+                if (colon == readme && repo->defbranch)
+                        *ref = xstrdup(repo->defbranch);
+                else
+                        *ref = xstrndup(readme, colon - readme);
+                readme = colon + 1;
+        }
+
+        /* Prepend repo path to relative readme path unless tracked. */
+        if (!(*ref) && readme[0] != '/')
+                *filename = fmtalloc("%s/%s", repo->path, readme);
+        else
+                *filename = xstrdup(readme);
+}
 static void choose_readme(struct cgit_repo *repo)
 {
-        char *entry, *filename, *ref;
+        int found;
+        char *filename, *ref;
+        struct string_list_item *entry;
 
-        /* If there's no space, we skip the possibly expensive
-         * selection process. */
-        if (!repo->readme || !strchr(repo->readme, ' '))
+        if (!repo->readme.nr)
                 return;
 
-        for (entry = strtok(repo->readme, " "); entry; entry = strtok(NULL, " ")) {
-                cgit_parse_readme(entry, NULL, &filename, &ref, repo);
-                if (!(*filename)) {
+        found = 0;
+        for_each_string_list_item(entry, &repo->readme) {
+                parse_readme(entry->string, &filename, &ref, repo);
+                if (!filename) {
                         free(filename);
                         free(ref);
                         continue;
                 }
-                if (*ref && cgit_ref_path_exists(filename, ref)) {
-                        free(filename);
-                        free(ref);
+                /* If there's only one item, we skip the possibly expensive
+                 * selection process. */
+                if (repo->readme.nr == 1) {
+                        found = 1;
                         break;
                 }
-                if (!access(filename, R_OK)) {
-                        free(filename);
-                        free(ref);
+                if (ref) {
+                        if (cgit_ref_path_exists(filename, ref, 1)) {
+                                found = 1;
+                                break;
+                        }
+                }
+                else if (!access(filename, R_OK)) {
+                        found = 1;
                         break;
                 }
                 free(filename);
                 free(ref);
         }
-        repo->readme = entry;
+        repo->readme.strdup_strings = 1;
+        string_list_clear(&repo->readme, 0);
+        repo->readme.strdup_strings = 0;
+        if (found)
+                string_list_append(&repo->readme, filename)->util = ref;
 }
 
 static int prepare_repo_cmd(struct cgit_context *ctx)
@@ -660,6 +701,7 @@ static char *get_first_line(char *txt)
 
 static void print_repo(FILE *f, struct cgit_repo *repo)
 {
+        struct string_list_item *item;
         fprintf(f, "repo.url=%s\n", repo->url);
         fprintf(f, "repo.name=%s\n", repo->name);
         fprintf(f, "repo.path=%s\n", repo->path);
@@ -670,8 +712,12 @@ static void print_repo(FILE *f, struct cgit_repo *repo)
                 fprintf(f, "repo.desc=%s\n", tmp);
                 free(tmp);
         }
-        if (repo->readme)
-                fprintf(f, "repo.readme=%s\n", repo->readme);
+        for_each_string_list_item(item, &repo->readme) {
+                if (item->util)
+                        fprintf(f, "repo.readme=%s:%s\n", (char *)item->util, item->string);
+                else
+                        fprintf(f, "repo.readme=%s\n", item->string);
+        }
         if (repo->defbranch)
                 fprintf(f, "repo.defbranch=%s\n", repo->defbranch);
         if (repo->module_link)
diff --git a/cgit.h b/cgit.h
index 850b792..f28cf30 100644
--- a/cgit.h
+++ b/cgit.h
@@ -73,7 +73,7 @@ struct cgit_repo {
         char *owner;
         char *defbranch;
         char *module_link;
-        char *readme;
+        struct string_list readme;
         char *section;
         char *clone_url;
         char *logo;
@@ -183,7 +183,7 @@ struct cgit_config {
         char *mimetype_file;
         char *module_link;
         char *project_list;
-        char *readme;
+        struct string_list readme;
         char *robots;
         char *root_title;
         char *root_desc;
diff --git a/cgitrc.5.txt b/cgitrc.5.txt
index 12a843b..6437ef4 100644
--- a/cgitrc.5.txt
+++ b/cgitrc.5.txt
@@ -290,8 +290,9 @@ project-list::
 
 readme::
         Text which will be used as default value for "repo.readme". Multiple
-        files may be specified, separated by a space, and cgit will use the
-        first found file in this list. Default value: none.
+        config keys may be specified, and cgit will use the first found file
+        in this list. This is useful in conjunction with scan-path. Default
+        value: none. See also: scan-path, repo.readme.
 
 remove-suffix::
         If set to "1" and scan-path is enabled, if any repositories are found
diff --git a/cmd.c b/cmd.c
index 198bf2f..abe8e46 100644
--- a/cmd.c
+++ b/cmd.c
@@ -1,6 +1,7 @@
 /* cmd.c: the cgit command dispatcher
  *
  * Copyright (C) 2008 Lars Hjemli
+ * Copyright (C) 2013 Jason A. Donenfeld <Jason@zx2c4.com>.
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
@@ -46,7 +47,7 @@ static void about_fn(struct cgit_context *ctx)
 
 static void blob_fn(struct cgit_context *ctx)
 {
-        cgit_print_blob(ctx->qry.sha1, ctx->qry.path, ctx->qry.head);
+        cgit_print_blob(ctx->qry.sha1, ctx->qry.path, ctx->qry.head, 0);
 }
 
 static void commit_fn(struct cgit_context *ctx)
diff --git a/scan-tree.c b/scan-tree.c
index a1ec8fb..2684b44 100644
--- a/scan-tree.c
+++ b/scan-tree.c
@@ -1,7 +1,7 @@
 /* scan-tree.c
  * 
  * Copyright (C) 2008-2009 Lars Hjemli
- * Copyright (C) 2010, 2012 Jason A. Donenfeld <Jason@zx2c4.com>
+ * Copyright (C) 2010-2013 Jason A. Donenfeld <Jason@zx2c4.com>
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
@@ -147,12 +147,6 @@ static void add_repo(const char *base, struct strbuf *path, repo_config_fn fn)
                 strbuf_setlen(path, pathlen);
         }
 
-        if (!repo->readme) {
-                strbuf_addstr(path, "README.html");
-                if (!stat(path->buf, &st))
-                        repo->readme = "README.html";
-                strbuf_setlen(path, pathlen);
-        }
         if (ctx.cfg.section_from_path) {
                 n  = ctx.cfg.section_from_path;
                 if (n > 0) {
diff --git a/ui-blob.c b/ui-blob.c
index b4be139..eb14a75 100644
--- a/ui-blob.c
+++ b/ui-blob.c
@@ -1,7 +1,7 @@
 /* ui-blob.c: show blob content
  *
  * Copyright (C) 2008 Lars Hjemli
- * Copyright (C) 2010 Jason A. Donenfeld <Jason@zx2c4.com>
+ * Copyright (C) 2010-2013 Jason A. Donenfeld <Jason@zx2c4.com>
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
@@ -15,7 +15,8 @@
 struct walk_tree_context {
         const char *match_path;
         unsigned char *matched_sha1;
-        int found_path;
+        int found_path:1;
+        int file_only:1;
 };
 
 static int walk_tree(const unsigned char *sha1, const char *base, int baselen,
@@ -23,6 +24,8 @@ static int walk_tree(const unsigned char *sha1, const char *base, int baselen,
 {
         struct walk_tree_context *walk_tree_ctx = cbdata;
 
+        if (walk_tree_ctx->file_only && !S_ISREG(mode))
+                return READ_TREE_RECURSIVE;
         if (strncmp(base, walk_tree_ctx->match_path, baselen)
                 || strcmp(walk_tree_ctx->match_path + baselen, pathname))
                 return READ_TREE_RECURSIVE;
@@ -31,33 +34,34 @@ static int walk_tree(const unsigned char *sha1, const char *base, int baselen,
         return 0;
 }
 
-int cgit_ref_path_exists(const char *path, const char *ref)
+int cgit_ref_path_exists(const char *path, const char *ref, int file_only)
 {
-        unsigned char sha1[20];
-        unsigned long size;
-        struct pathspec_item path_items = {
-                .match = path,
-                .len = strlen(path)
-        };
-        struct pathspec paths = {
-                .nr = 1,
-                .items = &path_items
-        };
-        struct walk_tree_context walk_tree_ctx = {
-                .match_path = path,
-                .matched_sha1 = sha1,
-                .found_path = 0
-        };
+        unsigned char sha1[20];
+        unsigned long size;
+        struct pathspec_item path_items = {
+                .match = path,
+                .len = strlen(path)
+        };
+        struct pathspec paths = {
+                .nr = 1,
+                .items = &path_items
+        };
+        struct walk_tree_context walk_tree_ctx = {
+                .match_path = path,
+                .matched_sha1 = sha1,
+                .found_path = 0,
+                .file_only = file_only
+        };
 
-        if (get_sha1(ref, sha1))
-                return 0;
-        if (sha1_object_info(sha1, &size) != OBJ_COMMIT) 
-                return 0;
-        read_tree_recursive(lookup_commit_reference(sha1)->tree, "", 0, 0, &paths, walk_tree, &walk_tree_ctx);
-        return walk_tree_ctx.found_path;
+        if (get_sha1(ref, sha1))
+                return 0;
+        if (sha1_object_info(sha1, &size) != OBJ_COMMIT) 
+                return 0;
+        read_tree_recursive(lookup_commit_reference(sha1)->tree, "", 0, 0, &paths, walk_tree, &walk_tree_ctx);
+        return walk_tree_ctx.found_path;
 }
 
-int cgit_print_file(char *path, const char *head)
+int cgit_print_file(char *path, const char *head, int file_only)
 {
         unsigned char sha1[20];
         enum object_type type;
@@ -75,7 +79,8 @@ int cgit_print_file(char *path, const char *head)
         struct walk_tree_context walk_tree_ctx = {
                 .match_path = path,
                 .matched_sha1 = sha1,
-                .found_path = 0
+                .found_path = 0,
+                .file_only = file_only
         };
 
         if (get_sha1(head, sha1))
@@ -98,7 +103,7 @@ int cgit_print_file(char *path, const char *head)
         return 0;
 }
 
-void cgit_print_blob(const char *hex, char *path, const char *head)
+void cgit_print_blob(const char *hex, char *path, const char *head, int file_only)
 {
         unsigned char sha1[20];
         enum object_type type;
@@ -116,6 +121,8 @@ void cgit_print_blob(const char *hex, char *path, const char *head)
         struct walk_tree_context walk_tree_ctx = {
                 .match_path = path,
                 .matched_sha1 = sha1,
+                .found_path = 0,
+                .file_only = file_only
         };
 
         if (hex) {
diff --git a/ui-blob.h b/ui-blob.h
index ce3649f..16847b2 100644
--- a/ui-blob.h
+++ b/ui-blob.h
@@ -1,8 +1,8 @@
 #ifndef UI_BLOB_H
 #define UI_BLOB_H
 
-extern int cgit_ref_path_exists(const char *path, const char *ref);
-extern int cgit_print_file(char *path, const char *head);
-extern void cgit_print_blob(const char *hex, char *path, const char *head);
+extern int cgit_ref_path_exists(const char *path, const char *ref, int file_only);
+extern int cgit_print_file(char *path, const char *head, int file_only);
+extern void cgit_print_blob(const char *hex, char *path, const char *head, int file_only);
 
 #endif /* UI_BLOB_H */
diff --git a/ui-shared.c b/ui-shared.c
index 519eef7..7ab2ab1 100644
--- a/ui-shared.c
+++ b/ui-shared.c
@@ -864,7 +864,7 @@ void cgit_print_pageheader(struct cgit_context *ctx)
                 if (ctx->repo->max_stats)
                         cgit_stats_link("stats", NULL, hc(ctx, "stats"),
                                         ctx->qry.head, ctx->qry.vpath);
-                if (ctx->repo->readme)
+                if (ctx->repo->readme.nr)
                         reporevlink("about", "about", NULL,
                                     hc(ctx, "about"), ctx->qry.head, NULL,
                                     NULL);
diff --git a/ui-summary.c b/ui-summary.c
index 57206dd..d8500d6 100644
--- a/ui-summary.c
+++ b/ui-summary.c
@@ -1,7 +1,7 @@
 /* ui-summary.c: functions for generating repo summary page
  *
  * Copyright (C) 2006 Lars Hjemli
- * Copyright (C) 2010 Jason A. Donenfeld <Jason@zx2c4.com>
+ * Copyright (C) 2010-2013 Jason A. Donenfeld <Jason@zx2c4.com>
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
@@ -13,6 +13,7 @@
 #include "ui-log.h"
 #include "ui-refs.h"
 #include "ui-blob.h"
+#include <libgen.h>
 
 static void print_url(char *base, char *suffix)
 {
@@ -95,69 +96,57 @@ void cgit_print_summary()
         html("</table>");
 }
 
-/* The caller must free filename and ref after calling this. */
-void cgit_parse_readme(const char *readme, const char *path, char **filename, char **ref, struct cgit_repo *repo)
+/* The caller must free the return value. */
+static char* append_readme_path(const char *filename, const char *ref, const char *path)
 {
-        const char *slash, *colon;
-        char *resolved_base, *resolved_full;
+        char *file, *base_dir, *full_path, *resolved_base = NULL, *resolved_full = NULL;