1 file changed, 136 insertions, 118 deletions

diff --git a/source/scan-tree.c b/source/scan-tree.c
index 89f65c9..4a1444f 100644
--- a/source/scan-tree.c
+++ b/source/scan-tree.c
@@ -1,86 +1,120 @@
-/* scan-tree.c
- *
- * Copyright (C) 2006-2014 cgit Development Team <cgit@lists.zx2c4.com>
- *
- * Licensed under GNU General Public License v2
- *   (see LICENSE.txt for full license text)
+/*
+ * Discovery of repositories on disk, so that an administrator can point cgit
+ * at a directory instead of naming every repository in cgitrc. The tree below
+ * that directory is walked for bare repositories and working trees, or when a
+ * project list is configured only the paths it names are visited. Every
+ * repository found is registered under its path relative to the root of the
+ * scan, with owner, description and section taken from the files git keeps
+ * beside it, and a cgitrc in the repository has the last word over all of them.
  */
 
 #include "cgit.h"
+#include "config.h"
 #include "scan-tree.h"
-#include "configfile.h"
-#include "html.h"
-#include <config.h>
+#include "shared.h"
 
-// The description git puts in every freshly created repository.
+// Git writes this line into the description file of every repository it
+// creates, so a repository still carrying it gets cgit's own default instead.
 static const char *default_git_desc = "Unnamed repository; edit this file 'description' to name the repository.";
 
-static struct cgit_repo *repo;
+// The config callbacks are handed only a name and a value, so the repository
+// being filled in waits here for the length of one add_repo call.
+static struct cgit_repo *current_repo;
 
-/* return 1 if path contains a objects/ directory and a HEAD file */
-static int is_git_dir(const char *path)
+static int stat_entry(const char *dir, const char *name, struct stat *st)
 {
-	struct stat st;
-	struct strbuf pathbuf = STRBUF_INIT;
-	int result = 0;
-
-	strbuf_addf(&pathbuf, "%s/objects", path);
-	if (stat(pathbuf.buf, &st)) {
-		if (errno != ENOENT)
-			fprintf(stderr, "Error checking path %s: %s (%d)\n",
-				path, strerror(errno), errno);
-		goto out;
-	}
-	if (!S_ISDIR(st.st_mode))
-		goto out;
-
-	strbuf_reset(&pathbuf);
-	strbuf_addf(&pathbuf, "%s/HEAD", path);
-	if (stat(pathbuf.buf, &st)) {
-		if (errno != ENOENT)
-			fprintf(stderr, "Error checking path %s: %s (%d)\n",
-				path, strerror(errno), errno);
-		goto out;
-	}
-	if (!S_ISREG(st.st_mode))
-		goto out;
+	struct strbuf path = STRBUF_INIT;
+	int err;
 
-	result = 1;
-out:
-	strbuf_release(&pathbuf);
-	return result;
+	strbuf_addf(&path, "%s/%s", dir, name);
+	err = stat(path.buf, st);
+	// A missing entry is the ordinary answer for a directory that is not a
+	// repository, so only some other failure is worth reporting.
+	if (err && errno != ENOENT)
+		fprintf(stderr, "Error checking path %s: %s (%d)\n",
+			dir, strerror(errno), errno);
+	strbuf_release(&path);
+	return err;
 }
 
-static void scan_tree_repo_config(const char *name, const char *value)
+static int is_git_dir(const char *path)
 {
-	cgit_repo_config(repo, name, value);
+	struct stat st;
+
+	if (stat_entry(path, "objects", &st) || !S_ISDIR(st.st_mode))
+		return 0;
+	if (stat_entry(path, "HEAD", &st) || !S_ISREG(st.st_mode))
+		return 0;
+	return 1;
 }
 
-static int gitconfig_config(const char *key, const char *value,
+static int apply_gitconfig(const char *key, const char *value,
 		const __attribute__((unused)) struct config_context *cfg_ctx,
 		void *cb)
 {
 	const char *name;
 
 	if (!strcmp(key, "gitweb.owner"))
-		cgit_repo_config(repo, "owner", value);
+		cgit_repo_config(current_repo, "owner", value);
 	else if (!strcmp(key, "gitweb.description"))
-		cgit_repo_config(repo, "desc", value);
+		cgit_repo_config(current_repo, "desc", value);
 	else if (!strcmp(key, "gitweb.category"))
-		cgit_repo_config(repo, "section", value);
+		cgit_repo_config(current_repo, "section", value);
 	else if (!strcmp(key, "gitweb.homepage"))
-		cgit_repo_config(repo, "homepage", value);
+		cgit_repo_config(current_repo, "homepage", value);
 	else if (skip_prefix(key, "cgit.", &name))
-		cgit_repo_config(repo, name, value);
+		cgit_repo_config(current_repo, name, value);
 
 	return 0;
 }
 
-static char *xstrrchr(char *s, char *from, int c)
+static void apply_cgitrc(const char *name, const char *value)
 {
-	while (from >= s && *from != c)
+	cgit_repo_config(current_repo, name, value);
+}
+
+static char *find_char_back(char *start, char *from, int c)
+{
+	while (from >= start && *from != c)
 		from--;
-	return from < s ? NULL : from;
+	return from < start ? NULL : from;
+}
+
+/*
+ * A positive depth counts separators from the left of the path and a negative
+ * one counts back from the right.
+ */
+static char *section_slash(struct strbuf *relpath, int depth)
+{
+	char *slash;
+
+	if (depth > 0) {
+		slash = relpath->buf - 1;
+		while (slash && depth && (slash = strchr(slash + 1, '/')))
+			depth--;
+	} else {
+		slash = relpath->buf + relpath->len;
+		while (slash && depth &&
+		       (slash = find_char_back(relpath->buf, slash - 1, '/')))
+			depth++;
+	}
+	return slash && !depth ? slash : NULL;
+}
+
+static void set_section_from_path(struct strbuf *relpath, int depth)
+{
+	char *slash = section_slash(relpath, depth);
+
+	if (!slash)
+		return;
+	*slash = '\0';
+	current_repo->section = cgit_strdup_first_line(relpath->buf);
+	*slash = '/';
+	if (starts_with(current_repo->name, current_repo->section)) {
+		current_repo->name += strlen(current_repo->section);
+		if (*current_repo->name == '/')
+			current_repo->name++;
+	}
 }
 
 static void add_repo(const char *base, struct strbuf *path)
@@ -88,10 +122,9 @@ static void add_repo(const char *base, struct strbuf *path)
 	struct stat st;
 	struct passwd *pwd;
 	size_t pathlen;
-	struct strbuf rel = STRBUF_INIT;
-	char *p, *slash;
-	int n;
-	size_t size;
+	struct strbuf relpath = STRBUF_INIT;
+	char *comma;
+	size_t desc_size;
 
 	if (stat(path->buf, &st)) {
 		fprintf(stderr, "Error accessing %s: %s (%d)\n",
@@ -104,7 +137,7 @@ static void add_repo(const char *base, struct strbuf *path)
 
 	if (ctx.cfg.strict_export) {
 		strbuf_addstr(path, ctx.cfg.strict_export);
-		if(stat(path->buf, &st))
+		if (stat(path->buf, &st))
 			return;
 		strbuf_setlen(path, pathlen);
 	}
@@ -115,86 +148,79 @@ static void add_repo(const char *base, struct strbuf *path)
 	strbuf_setlen(path, pathlen);
 
 	if (!starts_with(path->buf, base))
-		strbuf_addbuf(&rel, path);
+		strbuf_addbuf(&relpath, path);
 	else
-		strbuf_addstr(&rel, path->buf + strlen(base) + 1);
+		strbuf_addstr(&relpath, path->buf + strlen(base) + 1);
 
 	// Drop the trailing slash added above before testing for "/.git", since
 	// with it still attached the suffix never matches and every ordinary
 	// working tree is named "repo/.git" instead of "repo".
-	if (rel.len && rel.buf[rel.len - 1] == '/')
-		strbuf_setlen(&rel, rel.len - 1);
-	if (rel.len >= 5 && !strcmp(rel.buf + rel.len - 5, "/.git"))
-		strbuf_setlen(&rel, rel.len - 5);
+	if (relpath.len && relpath.buf[relpath.len - 1] == '/')
+		strbuf_setlen(&relpath, relpath.len - 1);
+	if (relpath.len >= 5 && !strcmp(relpath.buf + relpath.len - 5, "/.git"))
+		strbuf_setlen(&relpath, relpath.len - 5);
 
-	repo = cgit_add_repo(rel.buf);
+	current_repo = cgit_add_repo(relpath.buf);
 	if (ctx.cfg.enable_git_config) {
 		strbuf_addstr(path, "config");
-		git_config_from_file(gitconfig_config, path->buf, NULL);
+		git_config_from_file(apply_gitconfig, path->buf, NULL);
 		strbuf_setlen(path, pathlen);
 	}
 
 	if (ctx.cfg.remove_suffix) {
 		size_t urllen;
-		strip_suffix(repo->url, ".git", &urllen);
-		strip_suffix_mem(repo->url, &urllen, "/");
-		repo->url[urllen] = '\0';
+		strip_suffix(current_repo->url, ".git", &urllen);
+		strip_suffix_mem(current_repo->url, &urllen, "/");
+		current_repo->url[urllen] = '\0';
 	}
-	repo->path = cgit_strdup_first_line(path->buf);
-	while (!repo->owner) {
+	current_repo->path = cgit_strdup_first_line(path->buf);
+	while (!current_repo->owner) {
 		if ((pwd = getpwuid(st.st_uid)) == NULL) {
 			fprintf(stderr, "Error reading owner-info for %s: %s (%d)\n",
 				path->buf, strerror(errno), errno);
 			break;
 		}
+		// A gecos field puts the owner's name in front of a comma
+		// separated list of office and phone details.
 		if (pwd->pw_gecos)
-			if ((p = strchr(pwd->pw_gecos, ',')))
-				*p = '\0';
-		repo->owner = cgit_strdup_first_line(pwd->pw_gecos ? pwd->pw_gecos : pwd->pw_name);
+			if ((comma = strchr(pwd->pw_gecos, ',')))
+				*comma = '\0';
+		current_repo->owner = cgit_strdup_first_line(
+			pwd->pw_gecos ? pwd->pw_gecos : pwd->pw_name);
 	}
 
-	if (repo->desc == cgit_default_repo_desc || !repo->desc) {
+	if (current_repo->desc == cgit_default_repo_desc || !current_repo->desc) {
 		strbuf_addstr(path, "description");
 		if (!stat(path->buf, &st))
-			cgit_read_first_line(path->buf, &repo->desc, &size);
+			cgit_read_first_line(path->buf, &current_repo->desc,
+					     &desc_size);
 		strbuf_setlen(path, pathlen);
-		// Git writes this line into every repository it creates, so it
-		// describes nothing. Treat it as no description at all rather
-		// than repeating it down the whole index.
-		if (repo->desc && !strcmp(repo->desc, default_git_desc)) {
-			free(repo->desc);
-			repo->desc = cgit_default_repo_desc;
+		if (current_repo->desc &&
+		    !strcmp(current_repo->desc, default_git_desc)) {
+			free(current_repo->desc);
+			current_repo->desc = cgit_default_repo_desc;
 		}
 	}
 
-	if (ctx.cfg.section_from_path) {
-		n = ctx.cfg.section_from_path;
-		if (n > 0) {
-			slash = rel.buf - 1;
-			while (slash && n && (slash = strchr(slash + 1, '/')))
-				n--;
-		} else {
-			slash = rel.buf + rel.len;
-			while (slash && n && (slash = xstrrchr(rel.buf, slash - 1, '/')))
-				n++;
-		}
-		if (slash && !n) {
-			*slash = '\0';
-			repo->section = cgit_strdup_first_line(rel.buf);
-			*slash = '/';
-			if (starts_with(repo->name, repo->section)) {
-				repo->name += strlen(repo->section);
-				if (*repo->name == '/')
-					repo->name++;
-			}
-		}
-	}
+	if (ctx.cfg.section_from_path)
+		set_section_from_path(&relpath, ctx.cfg.section_from_path);
 
 	strbuf_addstr(path, "cgitrc");
 	if (!stat(path->buf, &st))
-		parse_configfile(path->buf, &scan_tree_repo_config);
+		config_file_parse(path->buf, &apply_cgitrc);
 
-	strbuf_release(&rel);
+	strbuf_release(&relpath);
+}
+
+static int should_scan(const struct dirent *ent)
+{
+	if (ent->d_name[0] != '.')
+		return 1;
+	if (ent->d_name[1] == '\0')
+		return 0;
+	if (ent->d_name[1] == '.' && ent->d_name[2] == '\0')
+		return 0;
+	return ctx.cfg.scan_hidden_path;
 }
 
 static void scan_path(const char *base, const char *path)
@@ -211,7 +237,7 @@ static void scan_path(const char *base, const char *path)
 		return;
 	}
 
-	strbuf_add(&pathbuf, path, strlen(path));
+	strbuf_add(&pathbuf, path, pathlen);
 	if (is_git_dir(pathbuf.buf)) {
 		add_repo(base, &pathbuf);
 		goto end;
@@ -221,20 +247,12 @@ static void scan_path(const char *base, const char *path)
 		add_repo(base, &pathbuf);
 		goto end;
 	}
-	/*
-	 * Add one because we don't want to lose the trailing '/' when we
-	 * reset the length of pathbuf in the loop below.
-	 */
+	// Take in the '/' that "/.git" left in the buffer, since the loop below
+	// truncates to this length and then appends an entry name straight on.
 	pathlen++;
 	while ((ent = readdir(dir)) != NULL) {
-		if (ent->d_name[0] == '.') {
-			if (ent->d_name[1] == '\0')
-				continue;
-			if (ent->d_name[1] == '.' && ent->d_name[2] == '\0')
-				continue;
-			if (!ctx.cfg.scan_hidden_path)
-				continue;
-		}
+		if (!should_scan(ent))
+			continue;
 		strbuf_setlen(&pathbuf, pathlen);
 		strbuf_addstr(&pathbuf, ent->d_name);
 		if (stat(pathbuf.buf, &st)) {