Use utf8::reencode_string from git

This replaces the iconv-support in cgit with similar functions already
existing in git.

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

2 files changed, 5 insertions, 60 deletions

diff --git a/cgit.h b/cgit.h
index 6c7a941..6291c58 100644
--- a/cgit.h
+++ b/cgit.h
@@ -1,289 +1,290 @@
 #ifndef CGIT_H
 #define CGIT_H
 
 
 #include <git-compat-util.h>
 #include <cache.h>
 #include <grep.h>
 #include <object.h>
 #include <tree.h>
 #include <commit.h>
 #include <tag.h>
 #include <diff.h>
 #include <diffcore.h>
 #include <refs.h>
 #include <revision.h>
 #include <log-tree.h>
 #include <archive.h>
 #include <xdiff/xdiff.h>
+#include <utf8.h>
 
 
 /*
  * The valid cgit repo-commands
  */
 #define CMD_LOG      1
 #define CMD_COMMIT   2
 #define CMD_DIFF     3
 #define CMD_TREE     4
 #define CMD_BLOB     5
 #define CMD_SNAPSHOT 6
 #define CMD_TAG      7
 #define CMD_REFS     8
 
 /*
  * Dateformats used on misc. pages
  */
 #define FMT_LONGDATE "%Y-%m-%d %H:%M:%S"
 #define FMT_SHORTDATE "%Y-%m-%d"
 
 
 /*
  * Limits used for relative dates
  */
 #define TM_MIN    60
 #define TM_HOUR  (TM_MIN * 60)
 #define TM_DAY   (TM_HOUR * 24)
 #define TM_WEEK  (TM_DAY * 7)
 #define TM_YEAR  (TM_DAY * 365)
 #define TM_MONTH (TM_YEAR / 12.0)
 
 
 /*
  * Default encoding
  */
 #define PAGE_ENCODING "UTF-8"
 
 typedef void (*configfn)(const char *name, const char *value);
 typedef void (*filepair_fn)(struct diff_filepair *pair);
 typedef void (*linediff_fn)(char *line, int len);
 
 struct cacheitem {
 	char *name;
 	struct stat st;
 	int ttl;
 	int fd;
 };
 
 struct repoinfo {
 	char *url;
 	char *name;
 	char *path;
 	char *desc;
 	char *owner;
 	char *defbranch;
 	char *group;
 	char *module_link;
 	char *readme;
 	int snapshots;
 	int enable_log_filecount;
 	int enable_log_linecount;
 };
 
 struct repolist {
 	int length;
 	int count;
 	struct repoinfo *repos;
 };
 
 struct commitinfo {
 	struct commit *commit;
 	char *author;
 	char *author_email;
 	unsigned long author_date;
 	char *committer;
 	char *committer_email;
 	unsigned long committer_date;
 	char *subject;
 	char *msg;
 	char *msg_encoding;
 };
 
 struct taginfo {
 	char *tagger;
 	char *tagger_email;
 	int 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;
 };
 
 extern const char *cgit_version;
 
 extern struct repolist cgit_repolist;
 extern struct repoinfo *cgit_repo;
 extern int cgit_cmd;
 
 extern char *cgit_root_title;
 extern char *cgit_css;
 extern char *cgit_logo;
 extern char *cgit_index_header;
 extern char *cgit_index_info;
 extern char *cgit_logo_link;
 extern char *cgit_module_link;
 extern char *cgit_agefile;
 extern char *cgit_virtual_root;
 extern char *cgit_script_name;
 extern char *cgit_cache_root;
 extern char *cgit_repo_group;
 
 extern int cgit_nocache;
 extern int cgit_snapshots;
 extern int cgit_enable_index_links;
 extern int cgit_enable_log_filecount;
 extern int cgit_enable_log_linecount;
 extern int cgit_max_lock_attempts;
 extern int cgit_cache_root_ttl;
 extern int cgit_cache_repo_ttl;
 extern int cgit_cache_dynamic_ttl;
 extern int cgit_cache_static_ttl;
 extern int cgit_cache_max_create_time;
 extern int cgit_summary_log;
 extern int cgit_summary_tags;
 extern int cgit_summary_branches;
 
 extern int cgit_max_msg_len;
 extern int cgit_max_repodesc_len;
 extern int cgit_max_commit_count;
 
 extern int cgit_query_has_symref;
 extern int cgit_query_has_sha1;
 
 extern char *cgit_querystring;
 extern char *cgit_query_repo;
 extern char *cgit_query_page;
 extern char *cgit_query_search;
 extern char *cgit_query_grep;
 extern char *cgit_query_head;
 extern char *cgit_query_sha1;
 extern char *cgit_query_sha2;
 extern char *cgit_query_path;
 extern char *cgit_query_name;
 extern int   cgit_query_ofs;
 
 extern int htmlfd;
 
 extern int cgit_get_cmd_index(const char *cmd);
 extern struct repoinfo *cgit_get_repoinfo(const char *url);
 extern void cgit_global_config_cb(const char *name, const char *value);
 extern void cgit_repo_config_cb(const char *name, const char *value);
 extern void cgit_querystring_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 int hextoint(char c);
 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,
 			   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 void html(const char *txt);
 extern void htmlf(const char *format,...);
 extern void html_txt(char *txt);
 extern void html_ntxt(int len, char *txt);
 extern void html_attr(char *txt);
 extern void html_hidden(char *name, char *value);
 extern void html_option(char *value, char *text, char *selected_value);
 extern void html_link_open(char *url, char *title, char *class);
 extern void html_link_close(void);
 extern void html_filemode(unsigned short mode);
 extern int html_include(const char *filename);
 
 extern int cgit_read_config(const char *filename, configfn fn);
 extern int cgit_parse_query(char *txt, configfn fn);
 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 char *cache_safe_filename(const char *unsafe);
 extern int cache_lock(struct cacheitem *item);
 extern int cache_unlock(struct cacheitem *item);
 extern int cache_cancel_lock(struct cacheitem *item);
 extern int cache_exist(struct cacheitem *item);
 extern int cache_expired(struct cacheitem *item);
 
 extern char *cgit_repourl(const char *reponame);
 extern char *cgit_fileurl(const char *reponame, const char *pagename,
 			  const char *filename, const char *query);
 extern char *cgit_pageurl(const char *reponame, const char *pagename,
 			  const char *query);
 
 extern const char *cgit_repobasename(const char *reponame);
 
 extern void cgit_tree_link(char *name, char *title, char *class, char *head,
 			   char *rev, char *path);
 extern void cgit_log_link(char *name, char *title, char *class, char *head,
 			  char *rev, char *path, int ofs, char *grep,
 			  char *pattern);
 extern void cgit_commit_link(char *name, char *title, char *class, char *head,
 			     char *rev);
 extern void cgit_refs_link(char *name, char *title, char *class, char *head,
 			   char *rev, char *path);
 extern void cgit_snapshot_link(char *name, char *title, char *class,
 			       char *head, char *rev, char *archivename);
 extern void cgit_diff_link(char *name, char *title, char *class, char *head,
 			   char *new_rev, char *old_rev, char *path);
 
 extern void cgit_object_link(struct object *obj);
 
 extern void cgit_print_error(char *msg);
 extern void cgit_print_date(time_t secs, char *format);
 extern void cgit_print_age(time_t t, time_t max_relative, char *format);
 extern void cgit_print_docstart(char *title, struct cacheitem *item);
 extern void cgit_print_docend();
 extern void cgit_print_pageheader(char *title, int show_search);
 extern void cgit_print_snapshot_start(const char *mimetype,
 				      const char *filename,
 				      struct cacheitem *item);
 extern void cgit_print_branches(int maxcount);
 extern void cgit_print_tags(int maxcount);
 
 extern void cgit_print_repolist(struct cacheitem *item);
 extern void cgit_print_summary();
 extern void cgit_print_log(const char *tip, int ofs, int cnt, char *grep,
 			   char *pattern, char *path, int pager);
 extern void cgit_print_blob(struct cacheitem *item, const char *hex, char *path);
 extern void cgit_print_tree(const char *rev, char *path);
 extern void cgit_print_commit(char *hex);
 extern void cgit_print_refs();
 extern void cgit_print_tag(char *revname);
 extern void cgit_print_diff(const char *new_hex, const char *old_hex, const char *prefix);
 extern void cgit_print_snapshot(struct cacheitem *item, const char *head,
 				const char *hex, const char *prefix,
 				const char *filename, int snapshot);
 extern void cgit_print_snapshot_links(const char *repo, const char *head,
 				      const char *hex, int snapshots);
 extern int cgit_parse_snapshots_mask(const char *str);
 
 #endif /* CGIT_H */
diff --git a/parsing.c b/parsing.c
index c731084..e8c7ab9 100644
--- a/parsing.c
+++ b/parsing.c
@@ -1,388 +1,332 @@
 /* config.c: parsing of config files
  *
  * Copyright (C) 2006 Lars Hjemli
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
  */
 
-#include <iconv.h>
-
 #include "cgit.h"
 
 int next_char(FILE *f)
 {
 	int c = fgetc(f);
 	if (c=='\r') {
 		c = fgetc(f);
 		if (c!='\n') {
 			ungetc(c, f);
 			c = '\r';
 		}
 	}
 	return c;
 }
 
 void skip_line(FILE *f)
 {
 	int c;
 
 	while((c=next_char(f)) && c!='\n' && c!=EOF)
 		;
 }
 
 int read_config_line(FILE *f, char *line, const char **value, int bufsize)
 {
 	int i = 0, isname = 0;
 
 	*value = NULL;
 	while(i<bufsize-1) {
 		int c = next_char(f);
 		if (!isname && (c=='#' || c==';')) {
 			skip_line(f);
 			continue;
 		}
 		if (!isname && isspace(c))
 			continue;
 
 		if (c=='=' && !*value) {
 			line[i] = 0;
 			*value = &line[i+1];
 		} else if (c=='\n' && !isname) {
 			i = 0;
 			continue;
 		} else if (c=='\n' || c==EOF) {
 			line[i] = 0;
 			break;
 		} else {
 			line[i]=c;
 		}
 		isname = 1;
 		i++;
 	}
 	line[i+1] = 0;
 	return i;
 }
 
 int cgit_read_config(const char *filename, configfn fn)
 {
 	static int nesting;
 	int len;
 	char line[256];
 	const char *value;
 	FILE *f;
 
 	/* cancel deeply nested include-commands */
 	if (nesting > 8)
 		return -1;
 	if (!(f = fopen(filename, "r")))
 		return -1;
 	nesting++;
 	while((len = read_config_line(f, line, &value, sizeof(line))) > 0)
 		(*fn)(line, value);
 	nesting--;
 	fclose(f);
 	return 0;
 }
 
 char *convert_query_hexchar(char *txt)
 {
 	int d1, d2;
 	if (strlen(txt) < 3) {
 		*txt = '\0';
 		return txt-1;
 	}
 	d1 = hextoint(*(txt+1));
 	d2 = hextoint(*(txt+2));
 	if (d1<0 || d2<0) {
 		strcpy(txt, txt+3);
 		return txt-1;
 	} else {
 		*txt = d1 * 16 + d2;
 		strcpy(txt+1, txt+3);
 		return txt;
 	}
 }
 
 int cgit_parse_query(char *txt, configfn fn)
 {
 	char *t, *value = NULL, c;
 
 	if (!txt)
 		return 0;
 
 	t = txt = xstrdup(txt);
 
 	while((c=*t) != '\0') {
 		if (c=='=') {
 			*t = '\0';
 			value = t+1;
 		} else if (c=='+') {
 			*t = ' ';
 		} else if (c=='%') {
 			t = convert_query_hexchar(t);
 		} else if (c=='&') {
 			*t = '\0';
 			(*fn)(txt, value);
 			txt = t+1;
 			value = NULL;
 		}
 		t++;
 	}
 	if (t!=txt)
 		(*fn)(txt, value);
 	return 0;
 }
 
 /*
  * url syntax: [repo ['/' cmd [ '/' path]]]
  *   repo: any valid repo url, may contain '/'
  *   cmd:  log | commit | diff | tree | view | blob | snapshot
  *   path: any valid path, may contain '/'
  *
  */
 void cgit_parse_url(const char *url)
 {
 	char *cmd, *p;
 
 	cgit_repo = NULL;
 	if (!url || url[0] == '\0')
 		return;
 
 	cgit_repo = cgit_get_repoinfo(url);
 	if (cgit_repo) {
 		cgit_query_repo = cgit_repo->url;
 		return;
 	}
 
 	cmd = strchr(url, '/');
 	while (!cgit_repo && cmd) {
 		cmd[0] = '\0';
 		cgit_repo = cgit_get_repoinfo(url);
 		if (cgit_repo == NULL) {
 			cmd[0] = '/';
 			cmd = strchr(cmd + 1, '/');
 			continue;
 		}
 
 		cgit_query_repo = cgit_repo->url;
 		p = strchr(cmd + 1, '/');
 		if (p) {
 			p[0] = '\0';
 			if (p[1])
 				cgit_query_path = trim_end(p + 1, '/');
 		}
 		cgit_cmd = cgit_get_cmd_index(cmd + 1);
 		cgit_query_page = xstrdup(cmd + 1);
 		return;
 	}
 }
 
-static char *iconv_msg(char *msg, const char *encoding)
-{
-	iconv_t msg_conv = iconv_open(PAGE_ENCODING, encoding);
-	size_t inlen = strlen(msg);
-	char *in;
-	char *out;
-	size_t inleft;
-	size_t outleft;
-	char *buf;
-	char *ret;
-	size_t buf_sz;
-	int again, fail;
-
-	if(msg_conv == (iconv_t)-1)
-		return NULL;
-
-	buf_sz = inlen * 2;
-	buf = xmalloc(buf_sz+1);
-	do {
-		in = msg;
-		inleft = inlen;
-
-		out = buf;
-		outleft = buf_sz;
-		iconv(msg_conv, &in, &inleft, &out, &outleft);
-
-		if(inleft == 0) {
-			fail = 0;
-			again = 0;
-		} else if(inleft != 0 && errno == E2BIG) {
-			fail = 0;
-			again = 1;
-
-			buf_sz *= 2;
-			free(buf);
-			buf = xmalloc(buf_sz+1);
-		} else {
-			fail = 1;
-			again = 0;
-		}
-	} while(again && !fail);
-
-	if(fail) {
-		free(buf);
-		ret = NULL;
-	} else {
-		buf = xrealloc(buf, out - buf);
-		*out = 0;
-		ret = buf;
-	}
-
-	iconv_close(msg_conv);
-
-	return ret;
-}
-
 char *substr(const char *head, const char *tail)
 {
 	char *buf;
 
 	buf = xmalloc(tail - head + 1);
 	strncpy(buf, head, tail - head);
 	buf[tail - head] = '\0';
 	return buf;
 }
 
 struct commitinfo *cgit_parse_commit(struct commit *commit)
 {
 	struct commitinfo *ret;
 	char *p = commit->buffer, *t = commit->buffer;
 
 	ret = xmalloc(sizeof(*ret));
 	ret->commit = commit;
 	ret->author = NULL;
 	ret->author_email = NULL;
 	ret->committer = NULL;
 	ret->committer_email = NULL;
 	ret->subject = NULL;
 	ret->msg = NULL;
 	ret->msg_encoding = NULL;
 
 	if (p == NULL)
 		return ret;
 
 	if (strncmp(p, "tree ", 5))
 		die("Bad commit: %s", sha1_to_hex(commit->object.sha1));
 	else
 		p += 46; // "tree " + hex[40] + "\n"
 
 	while (!strncmp(p, "parent ", 7))
 		p += 48; // "parent " + hex[40] + "\n"
 
 	if (!strncmp(p, "author ", 7)) {
 		p += 7;
 		t = strchr(p, '<') - 1;
 		ret->author = substr(p, t);
 		p = t;
 		t = strchr(t, '>') + 1;
 		ret->author_email = substr(p, t);
 		ret->author_date = atol(++t);
 		p = strchr(t, '\n') + 1;
 	}
 
 	if (!strncmp(p, "committer ", 9)) {
 		p += 9;
 		t = strchr(p, '<') - 1;
 		ret->committer = substr(p, t);
 		p = t;
 		t = strchr(t, '>') + 1;
 		ret->committer_email = substr(p, t);
 		ret->committer_date = atol(++t);
 		p = strchr(t, '\n') + 1;
 	}
 
 	if (!strncmp(p, "encoding ", 9)) {
 		p += 9;
 		t = strchr(p, '\n') + 1;
 		ret->msg_encoding = substr(p, t);
 		p = t;
 	} else
 		ret->msg_encoding = xstrdup(PAGE_ENCODING);
 
 	while (*p && (*p != '\n'))
 		p = strchr(p, '\n') + 1; // skip unknown header fields
 
 	while (*p == '\n')
 		p = strchr(p, '\n') + 1;
 
 	t = strchr(p, '\n');
 	if (t) {
 		if (*t == '\0')
 			ret->subject = "** empty **";
 		else
 			ret->subject = substr(p, t);
 		p = t + 1;
 
 		while (*p == '\n')
 			p = strchr(p, '\n') + 1;
 		ret->msg = xstrdup(p);
 	} else
 		ret->subject = substr(p, p+strlen(p));
 
 	if(strcmp(ret->msg_encoding, PAGE_ENCODING)) {
-		t = iconv_msg(ret->subject, ret->msg_encoding);
+		t = reencode_string(ret->subject, PAGE_ENCODING,
+				    ret->msg_encoding);
 		if(t) {
 			free(ret->subject);
 			ret->subject = t;
 		}
 
-		t = iconv_msg(ret->msg, ret->msg_encoding);
+		t = reencode_string(ret->msg, PAGE_ENCODING,
+				    ret->msg_encoding);
 		if(t) {
 			free(ret->msg);
 			ret->msg = t;
 		}
 	}
 
 	return ret;
 }
 
 
 struct taginfo *cgit_parse_tag(struct tag *tag)
 {
 	void *data;
 	enum object_type type;
 	unsigned long size;
 	char *p, *t;
 	struct taginfo *ret;
 
 	data = read_sha1_file(tag->object.sha1, &type, &size);
 	if (!data || type != OBJ_TAG) {
 		free(data);
 		return 0;
 	}
 
 	ret = xmalloc(sizeof(*ret));
 	ret->tagger = NULL;
 	ret->tagger_email = NULL;
 	ret->tagger_date = 0;
 	ret->msg = NULL;
 
 	p = data;
 
 	while (p && *p) {
 		if (*p == '\n')
 			break;
 
 		if (!strncmp(p, "tagger ", 7)) {
 			p += 7;
 			t = strchr(p, '<') - 1;
 			ret->tagger = substr(p, t);
 			p = t;
 			t = strchr(t, '>') + 1;
 			ret->tagger_email = substr(p, t);
 			ret->tagger_date = atol(++t);
 		}
 		p = strchr(p, '\n') + 1;
 	}
 
 	while (p && *p && (*p != '\n'))
 		p = strchr(p, '\n') + 1; // skip unknown tag fields
 
 	while (p && (*p == '\n'))
 		p = strchr(p, '\n') + 1;
 	if (p && *p)
 		ret->msg = xstrdup(p);
 	free(data);
 	return ret;
 }