Convert subject and message with iconv_msg.

[ps-cgit] / parsing.c

diff --git a/parsing.c b/parsing.c
index a028625e23cafd7be713ede646eda35cfe6387a1..c7310847561d8d2a7edc11912c2991d12600fef9 100644 (file)
--- a/parsing.c
+++ b/parsing.c
@@ -6,6 +6,8 @@
  *   (see COPYING for full license text)
  */
 
  *   (see COPYING for full license text)
  */
 

+#include <iconv.h>
+

 #include "cgit.h"
 
 int next_char(FILE *f)
 #include "cgit.h"
 
 int next_char(FILE *f)


@@ -167,7 +169,8 @@ void cgit_parse_url(const char *url)
                p = strchr(cmd + 1, '/');
                if (p) {
                        p[0] = '\0';
                p = strchr(cmd + 1, '/');
                if (p) {
                        p[0] = '\0';

-                       cgit_query_path = xstrdup(p + 1);
+                       if (p[1])
+                               cgit_query_path = trim_end(p + 1, '/');

                }
                cgit_cmd = cgit_get_cmd_index(cmd + 1);
                cgit_query_page = xstrdup(cmd + 1);
                }
                cgit_cmd = cgit_get_cmd_index(cmd + 1);
                cgit_query_page = xstrdup(cmd + 1);


@@ -175,6 +178,62 @@ void cgit_parse_url(const char *url)
        }
 }
 
        }
 }
 

+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;
 char *substr(const char *head, const char *tail)
 {
        char *buf;


@@ -198,6 +257,10 @@ struct commitinfo *cgit_parse_commit(struct commit *commit)
        ret->committer_email = NULL;
        ret->subject = NULL;
        ret->msg = 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));
 
        if (strncmp(p, "tree ", 5))
                die("Bad commit: %s", sha1_to_hex(commit->object.sha1));


@@ -229,18 +292,48 @@ struct commitinfo *cgit_parse_commit(struct commit *commit)
                p = strchr(t, '\n') + 1;
        }
 
                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');
        while (*p == '\n')
                p = strchr(p, '\n') + 1;
 
        t = strchr(p, '\n');

-       if (t && *t) {
-               ret->subject = substr(p, t);
+       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;
                p = t + 1;
 
                while (*p == '\n')
                        p = strchr(p, '\n') + 1;

-               ret->msg = p;
+               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);
+               if(t) {
+                       free(ret->subject);
+                       ret->subject = t;
+               }
+
+               t = iconv_msg(ret->msg, ret->msg_encoding);
+               if(t) {
+                       free(ret->msg);
+                       ret->msg = t;
+               }

        }
        }

+

        return ret;
 }
 
        return ret;
 }
 


@@ -283,6 +376,9 @@ struct taginfo *cgit_parse_tag(struct tag *tag)
                p = strchr(p, '\n') + 1;
        }
 
                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)
        while (p && (*p == '\n'))
                p = strchr(p, '\n') + 1;
        if (p && *p)