Merge pull request #119 from adamemerson/siteconf-user-agent

Allow setting User Agent in Siteconf

4 files changed, 26 insertions, 2 deletions

diff --git a/doc/README.siteconf b/doc/README.siteconf
index 39b1028..5eb2554 100644
--- a/doc/README.siteconf
+++ b/doc/README.siteconf
@@ -13,6 +13,7 @@ substitute_url "<destination-url>"
 url_charset <charset>
 no_referer_from on|off
 no_referer_to on|off
+user_agent "string"
 
 The last match wins.
 
@@ -40,6 +41,12 @@ url_charset utf-8
 When combinated with "decode_url" option turned on, links to
 Wikipedia will be human-readable.
 
+url m@^https?://(.*\.)google\.com/@
+user_agent "Lynx/2.8.8dev.3 libwww-FM/2.14 SSL-MM/1.4.1"
+
+Tell Google we're actually Lynx. (So they send us a text-browser friendly
+results page.)
+
 ===== Regular expressions notes =====
 
 Following expressions are all equivalent:
diff --git a/fm.h b/fm.h
index 8face3c..4a17ecc 100644
--- a/fm.h
+++ b/fm.h
@@ -271,8 +271,10 @@ extern int REV_LB[];
 #define SCONF_URL_CHARSET	2
 #define SCONF_NO_REFERER_FROM	3
 #define SCONF_NO_REFERER_TO	4
-#define SCONF_N_FIELD		5
+#define SCONF_USER_AGENT	5
+#define SCONF_N_FIELD		6
 #define query_SCONF_SUBSTITUTE_URL(pu) ((const char *)querySiteconf(pu, SCONF_SUBSTITUTE_URL))
+#define query_SCONF_USER_AGENT(pu) ((const char *)querySiteconf(pu, SCONF_USER_AGENT))
 #define query_SCONF_URL_CHARSET(pu) ((const wc_ces *)querySiteconf(pu, SCONF_URL_CHARSET))
 #define query_SCONF_NO_REFERER_FROM(pu) ((const int *)querySiteconf(pu, SCONF_NO_REFERER_FROM))
 #define query_SCONF_NO_REFERER_TO(pu) ((const int *)querySiteconf(pu, SCONF_NO_REFERER_TO))
diff --git a/rc.c b/rc.c
index 3fd84ef..25f7a56 100644
--- a/rc.c
+++ b/rc.c
@@ -1602,6 +1602,7 @@ helpFile(char *base)
  * url_charset <charset>
  * no_referer_from on|off
  * no_referer_to on|off
+ * user_agent "<string>"
  * 
  * The last match wins.
  */
@@ -1614,6 +1615,7 @@ struct siteconf_rec {
     unsigned char mask[(SCONF_N_FIELD + 7) >> 3];
 
     char *substitute_url;
+    char *user_agent;
 #ifdef USE_M17N
     wc_ces url_charset;
 #endif
@@ -1640,6 +1642,7 @@ newSiteconfRec(void)
     memset(ent->mask, 0, sizeof(ent->mask));
 
     ent->substitute_url = NULL;
+    ent->user_agent = NULL;
 #ifdef USE_M17N
     ent->url_charset = 0;
 #endif
@@ -1718,6 +1721,10 @@ loadSiteconf(void)
 	    ent->substitute_url = getQWord(&p);
 	    SCONF_SET(ent, SCONF_SUBSTITUTE_URL);
 	}
+	if (strcmp(s, "user_agent") == 0) {
+	    ent->user_agent = getQWord(&p);
+	    SCONF_SET(ent, SCONF_USER_AGENT);
+	}
 #ifdef USE_M17N
 	else if (strcmp(s, "url_charset") == 0) {
 	    char *charset = getWord(&p);
@@ -1797,6 +1804,11 @@ url_found:
 	    return tmp->ptr;
 	}
 	return NULL;
+    case SCONF_USER_AGENT:
+	if (ent->user_agent && *ent->user_agent) {
+	    return ent->user_agent;
+	}
+	return NULL;
 #ifdef USE_M17N
     case SCONF_URL_CHARSET:
 	return &ent->url_charset;
diff --git a/url.c b/url.c
index 31d7c4b..aab7b9d 100644
--- a/url.c
+++ b/url.c
@@ -1323,10 +1323,13 @@ otherinfo(ParsedURL *target, ParsedURL *current, char *referer)
     Str s = Strnew();
     const int *no_referer_ptr;
     int no_referer;
+    const char* url_user_agent = query_SCONF_USER_AGENT(target);
 
     if (!override_user_agent) {
         Strcat_charp(s, "User-Agent: ");
-        if (UserAgent == NULL || *UserAgent == '\0')
+	if (url_user_agent)
+	   Strcat_charp(s, url_user_agent);
+	else if (UserAgent == NULL || *UserAgent == '\0')
             Strcat_charp(s, w3m_version);
         else
             Strcat_charp(s, UserAgent);