Subject: New feature: siteconf From: AIDA Shinra Date: Wed, 27 Jun 2012 20:43:46 +0900 Origin: upstream, http://www.sic.med.tohoku.ac.jp/~satodai/w3m-dev/201206.month/4464.html Patch to support the siteconf feature, from [w3m-dev 04464] on 2012-06-27, provided by AIDA Shinra. diff --git a/anchor.c b/anchor.c index 27bbd56..39f221d 100644 --- a/anchor.c +++ b/anchor.c @@ -200,10 +200,11 @@ _put_anchor_news(Buffer *buf, char *p1, char *p2, int line, int pos) if (*(p2 - 1) == '>') p2--; } - tmp = wc_Str_conv_strict(Strnew_charp_n(p1, p2 - p1), InnerCharset, - buf->document_charset); - tmp = Sprintf("news:%s", file_quote(tmp->ptr)); - return registerHref(buf, tmp->ptr, NULL, NO_REFERER, NULL, '\0', line, + tmp = Strnew_charp("news:"); + Strcat_charp_n(tmp, p1, p2 - p1); + return registerHref(buf, url_encode(tmp->ptr, baseURL(buf), + buf->document_charset), + NULL, NO_REFERER, NULL, '\0', line, pos); } #endif /* USE_NNTP */ @@ -213,9 +214,10 @@ _put_anchor_all(Buffer *buf, char *p1, char *p2, int line, int pos) { Str tmp; - tmp = wc_Str_conv_strict(Strnew_charp_n(p1, p2 - p1), InnerCharset, - buf->document_charset); - return registerHref(buf, url_quote(tmp->ptr), NULL, NO_REFERER, NULL, + tmp = Strnew_charp_n(p1, p2 - p1); + return registerHref(buf, url_encode(tmp->ptr, baseURL(buf), + buf->document_charset), + NULL, NO_REFERER, NULL, '\0', line, pos); } @@ -756,7 +758,7 @@ link_list_panel(Buffer *buf) p = parsedURL2Str(&pu)->ptr; u = html_quote(p); if (DecodeURL) - p = html_quote(url_unquote_conv(p, buf->document_charset)); + p = html_quote(url_decode2(p, buf)); else p = u; } @@ -787,7 +789,7 @@ link_list_panel(Buffer *buf) p = parsedURL2Str(&pu)->ptr; u = html_quote(p); if (DecodeURL) - p = html_quote(url_unquote_conv(p, buf->document_charset)); + p = html_quote(url_decode2(p, buf)); else p = u; t = getAnchorText(buf, al, a); @@ -809,16 +811,13 @@ link_list_panel(Buffer *buf) p = parsedURL2Str(&pu)->ptr; u = html_quote(p); if (DecodeURL) - p = html_quote(url_unquote_conv(p, buf->document_charset)); + p = html_quote(url_decode2(p, buf)); else p = u; if (a->title && *a->title) t = html_quote(a->title); - else if (DecodeURL) - t = html_quote(url_unquote_conv - (a->url, buf->document_charset)); else - t = html_quote(a->url); + t = html_quote(url_decode2(a->url, buf)); Strcat_m_charp(tmp, "
  • ", t, "
    ", p, "\n", NULL); a = retrieveAnchor(buf->formitem, a->start.line, a->start.pos); @@ -842,19 +841,13 @@ link_list_panel(Buffer *buf) p = parsedURL2Str(&pu)->ptr; u = html_quote(p); if (DecodeURL) - p = html_quote(url_unquote_conv(p, - buf-> - document_charset)); + p = html_quote(url_decode2(p, buf)); else p = u; if (m->alt && *m->alt) t = html_quote(m->alt); - else if (DecodeURL) - t = html_quote(url_unquote_conv(m->url, - buf-> - document_charset)); else - t = html_quote(m->url); + t = html_quote(url_decode2(m->url, buf)); Strcat_m_charp(tmp, "
  • ", t, "
    ", p, "\n", NULL); } diff --git a/config.h.in b/config.h.in index 2f41eed..59997b4 100644 --- a/config.h.in +++ b/config.h.in @@ -25,6 +25,7 @@ #define PASSWD_FILE RC_DIR "/passwd" #define PRE_FORM_FILE RC_DIR "/pre_form" +#define SITECONF_FILE RC_DIR "/siteconf" #define USER_MAILCAP RC_DIR "/mailcap" #define SYS_MAILCAP CONF_DIR "/mailcap" #define USER_MIMETYPES "~/.mime.types" diff --git a/display.c b/display.c index e00eb0c..2fe1183 100644 --- a/display.c +++ b/display.c @@ -257,7 +257,7 @@ make_lastline_link(Buffer *buf, char *title, char *url) parseURL2(url, &pu, baseURL(buf)); u = parsedURL2Str(&pu); if (DecodeURL) - u = Strnew_charp(url_unquote_conv(u->ptr, buf->document_charset)); + u = Strnew_charp(url_decode2(u->ptr, buf)); #ifdef USE_M17N u = checkType(u, &pr, NULL); #endif diff --git a/doc-jp/README.siteconf b/doc-jp/README.siteconf new file mode 100644 index 0000000..58b51c7 --- /dev/null +++ b/doc-jp/README.siteconf @@ -0,0 +1,60 @@ +siteconf: サイト別カスタマイズ + +siteconf は、 URL のパターンと、それに紐付けられた設定から成ります。 +siteconf を使うと、サイト毎に文字コードを指定して "decode_url" +の出力を改善したり、 Google のリダイレクタを迂回して性能や +プライバシーを向上させたりすることができます。 + +デフォルトでは siteconf は ~/.w3m/siteconf から読み込まれます。 + +===== 構文 ===== + +url |//|m@@i [exact] +substitute_url "" +url_charset +no_referer_from on|off +no_referer_to on|off + +後ろの方に書かれたものが優先されます。 + +===== 例 ===== + +url "http://twitter.com/#!/" +substitute_url "http://mobile.twitter.com/" + +twitter.com をモバイルサイトに転送します。 + +url "http://your.bookmark.net/" +no_referer_from on + +your.bookmark.net から張ったリンクを辿る際に、 HTTP referer を +送らないようにします。 + +url "http://www.google.com/url?" exact +substitute_url "file:///cgi-bin/your-redirector.cgi?" + +Google のリダイレクタを local CGI に転送します。 + +url /^http:\/\/[a-z]*\.wikipedia\.org\// +url_charset utf-8 + +同時に "decode_url" オプションをオンにすると、 Wikipedia への +リンクを UTF-8 としてデコードして表示します。 + +===== 正規表現について ===== + +次の正規表現はいずれも同じ意味を表します。 + +/http:\/\/www\.example\.com\// +m/http:\/\/www\.example\.com\// +m@http://www\.example\.com/@ +m!http://www\.example\.com/! + +最後に 'i' 修飾子を付けると、大文字小文字を区別せずに照合を行います。 +例えば、 m@^http://www\.example\.com/abc/@i は以下のいずれとも一致します。 + +http://www.example.com/abc/ +http://www.example.com/Abc/ +http://www.example.com/ABC/ + +ただし、ホスト名の部分は常に小文字に変換してから比較します。 diff --git a/doc/README.siteconf b/doc/README.siteconf new file mode 100644 index 0000000..f173087 --- /dev/null +++ b/doc/README.siteconf @@ -0,0 +1,60 @@ +The siteconf: Site-specific preferences + +The siteconf consists of URL patterns and preferences associated to them. +You can improve "decode_url" feature by giving charsets of URLs site by site, +or bypass Google's redirector for performance and your privacy. + +The siteconf is read from ~/.w3m/siteconf by default. + +===== The syntax ===== + +url |//|m@@i [exact] +substitute_url "" +url_charset +no_referer_from on|off +no_referer_to on|off + +The last match wins. + +===== Examples ===== + +url "http://twitter.com/#!/" +substitute_url "http://mobile.twitter.com/" + +This forwards the twitter.com to its mobile site. + +url "http://your.bookmark.net/" +no_referer_from on + +This prevents HTTP referers from being sent when you follow links +at the your.bookmark.net. + +url "http://www.google.com/url?" exact +substitute_url "file:///cgi-bin/your-redirector.cgi?" + +This forwards the Google's redirector to your local CGI. + +url /^http:\/\/[a-z]*\.wikipedia\.org\// +url_charset utf-8 + +When combinated with "decode_url" option turned on, links to +Wikipedia will be human-readable. + +===== Regular expressions notes ===== + +Following expressions are all equivalent: + +/http:\/\/www\.example\.com\// +m/http:\/\/www\.example\.com\// +m@http://www\.example\.com/@ +m!http://www\.example\.com/! + +With a trailing 'i' modifier, you can specify a case-insensitive match. +For example, m@^http://www\.example\.com/abc/@i matches to: + +http://www.example.com/abc/ +http://www.example.com/Abc/ +http://www.example.com/ABC/ + +Hostnames, however, are always converted to lowercases before compared. + diff --git a/file.c b/file.c index 567d41e..22b76d9 100644 --- a/file.c +++ b/file.c @@ -47,11 +47,11 @@ static JMP_BUF AbortLoading; static struct table *tables[MAX_TABLE]; static struct table_mode table_mode[MAX_TABLE]; -#ifdef USE_IMAGE +#if defined(USE_M17N) || defined(USE_IMAGE) static ParsedURL *cur_baseURL = NULL; -#ifdef USE_M17N -static char cur_document_charset; #endif +#ifdef USE_M17N +static wc_ces cur_document_charset = 0; #endif static Str cur_title; @@ -215,7 +215,6 @@ currentLn(Buffer *buf) static Buffer * loadSomething(URLFile *f, - char *path, Buffer *(*loadproc) (URLFile *, Buffer *), Buffer *defaultbuf) { Buffer *buf; @@ -223,17 +222,23 @@ loadSomething(URLFile *f, if ((buf = loadproc(f, defaultbuf)) == NULL) return NULL; - buf->filename = path; if (buf->buffername == NULL || buf->buffername[0] == '\0') { buf->buffername = checkHeader(buf, "Subject:"); - if (buf->buffername == NULL) - buf->buffername = conv_from_system(lastFileName(path)); + if (buf->buffername == NULL && buf->filename != NULL) + buf->buffername = conv_from_system(lastFileName(buf->filename)); } if (buf->currentURL.scheme == SCM_UNKNOWN) buf->currentURL.scheme = f->scheme; - buf->real_scheme = f->scheme; if (f->scheme == SCM_LOCAL && buf->sourcefile == NULL) - buf->sourcefile = path; + buf->sourcefile = buf->filename; + if (loadproc == loadHTMLBuffer +#ifdef USE_IMAGE + || loadproc == loadImageBuffer +#endif + ) + buf->type = "text/html"; + else + buf->type = "text/plain"; return buf; } @@ -484,28 +489,6 @@ convertLine0(URLFile *uf, Str line, int mode) return line; } -/* - * loadFile: load file to buffer - */ -Buffer * -loadFile(char *path) -{ - Buffer *buf; - URLFile uf; - init_stream(&uf, SCM_LOCAL, NULL); - examineFile(path, &uf); - if (uf.stream == NULL) - return NULL; - buf = newBuffer(INIT_BUFFER_WIDTH); - current_content_length = 0; -#ifdef USE_M17N - content_charset = 0; -#endif - buf = loadSomething(&uf, path, loadBuffer, buf); - UFclose(&uf); - return buf; -} - int matchattr(char *p, char *attr, int len, Str *value) { @@ -1697,13 +1680,15 @@ getLinkNumberStr(int correction) /* * loadGeneralFile: load file to buffer */ +#define DO_EXTERNAL ((Buffer *(*)(URLFile *, Buffer *))doExternal) Buffer * loadGeneralFile(char *path, ParsedURL *volatile current, char *referer, int flag, FormList *volatile request) { URLFile f, *volatile of = NULL; ParsedURL pu; - Buffer *b = NULL, *(*volatile proc)() = loadBuffer; + Buffer *b = NULL; + Buffer *(*volatile proc)(URLFile *, Buffer *) = loadBuffer; char *volatile tpath; char *volatile t = "text/plain", *p, *volatile real_type = NULL; Buffer *volatile t_buf = NULL; @@ -1730,7 +1715,22 @@ loadGeneralFile(char *path, ParsedURL *volatile current, char *referer, add_auth_cookie_flag = 0; checkRedirection(NULL); + load_doc: + { + const char *sc_redirect; + parseURL2(tpath, &pu, current); + sc_redirect = query_SCONF_SUBSTITUTE_URL(&pu); + if (sc_redirect && *sc_redirect && checkRedirection(&pu)) { + tpath = (char *)sc_redirect; + request = NULL; + add_auth_cookie_flag = 0; + current = New(ParsedURL); + *current = pu; + status = HTST_NORMAL; + goto load_doc; + } + } TRAP_OFF; url_option.referer = referer; url_option.flag = flag; @@ -1863,7 +1863,7 @@ loadGeneralFile(char *path, ParsedURL *volatile current, char *referer, /* 302: Found */ /* 303: See Other */ /* 307: Temporary Redirect (HTTP/1.1) */ - tpath = url_quote_conv(p, DocumentCharset); + tpath = url_encode(p, NULL, 0); request = NULL; UFclose(&f); current = New(ParsedURL); @@ -2022,7 +2022,7 @@ loadGeneralFile(char *path, ParsedURL *volatile current, char *referer, if (f.is_cgi && (p = checkHeader(t_buf, "Location:")) != NULL && checkRedirection(&pu)) { /* document moved */ - tpath = url_quote_conv(remove_space(p), DocumentCharset); + tpath = url_encode(remove_space(p), NULL, 0); request = NULL; UFclose(&f); add_auth_cookie_flag = 0; @@ -2123,10 +2123,6 @@ loadGeneralFile(char *path, ParsedURL *volatile current, char *referer, if (real_type == NULL) real_type = t; proc = loadBuffer; -#ifdef USE_IMAGE - cur_baseURL = New(ParsedURL); - copyParsedURL(cur_baseURL, &pu); -#endif current_content_length = 0; if ((p = checkHeader(t_buf, "Content-Length:")) != NULL) @@ -2197,18 +2193,8 @@ loadGeneralFile(char *path, ParsedURL *volatile current, char *referer, #endif else if (w3m_backend) ; else if (!(w3m_dump & ~DUMP_FRAME) || is_dump_text_type(t)) { - if (!do_download && doExternal(f, - pu.real_file ? pu.real_file : pu.file, - t, &b, t_buf)) { - if (b && b != NO_BUFFER) { - b->real_scheme = f.scheme; - b->real_type = real_type; - if (b->currentURL.host == NULL && b->currentURL.file == NULL) - copyParsedURL(&b->currentURL, &pu); - } - UFclose(&f); - TRAP_OFF; - return b; + if (!do_download && searchExtViewer(t) != NULL) { + proc = DO_EXTERNAL; } else { TRAP_OFF; @@ -2232,36 +2218,30 @@ loadGeneralFile(char *path, ParsedURL *volatile current, char *referer, else if (w3m_dump & DUMP_FRAME) return NULL; + if (t_buf == NULL) + t_buf = newBuffer(INIT_BUFFER_WIDTH); + copyParsedURL(&t_buf->currentURL, &pu); + t_buf->filename = pu.real_file ? pu.real_file : + pu.file ? conv_to_system(pu.file) : NULL; if (flag & RG_FRAME) { - if (t_buf == NULL) - t_buf = newBuffer(INIT_BUFFER_WIDTH); t_buf->bufferprop |= BP_FRAME; } #ifdef USE_SSL - if (t_buf) - t_buf->ssl_certificate = f.ssl_certificate; + t_buf->ssl_certificate = f.ssl_certificate; #endif frame_source = flag & RG_FRAME_SRC; - b = loadSomething(&f, pu.real_file ? pu.real_file : pu.file, proc, t_buf); + if (proc == DO_EXTERNAL) { + b = doExternal(f, t, t_buf); + } else { + b = loadSomething(&f, proc, t_buf); + } UFclose(&f); frame_source = 0; - if (b) { + if (b && b != NO_BUFFER) { b->real_scheme = f.scheme; b->real_type = real_type; - if (b->currentURL.host == NULL && b->currentURL.file == NULL) - copyParsedURL(&b->currentURL, &pu); - if (is_html_type(t)) - b->type = "text/html"; - else if (w3m_backend) { - Str s = Strnew_charp(t); - b->type = s->ptr; - } -#ifdef USE_IMAGE - else if (proc == loadImageBuffer) - b->type = "text/html"; -#endif - else - b->type = "text/plain"; + if (w3m_backend) + b->type = allocStr(t, -1); if (pu.label) { if (proc == loadHTMLBuffer) { Anchor *a; @@ -3228,7 +3208,7 @@ process_img(struct parsed_tag *tag, int width) if (!parsedtag_get_value(tag, ATTR_SRC, &p)) return tmp; - p = remove_space(p); + p = url_encode(remove_space(p), cur_baseURL, cur_document_charset); q = NULL; parsedtag_get_value(tag, ATTR_ALT, &q); if (!pseudoInlines && (q == NULL || (*q == '\0' && ignore_null_img_alt))) @@ -3322,12 +3302,7 @@ process_img(struct parsed_tag *tag, int width) Image image; ParsedURL u; -#ifdef USE_M17N - parseURL2(wc_conv(p, InnerCharset, cur_document_charset)->ptr, &u, - cur_baseURL); -#else parseURL2(p, &u, cur_baseURL); -#endif image.url = parsedURL2Str(&u)->ptr; if (!uncompressed_file_type(u.file, &image.ext)) image.ext = filename_extension(u.file, TRUE); @@ -4084,6 +4059,7 @@ process_form_int(struct parsed_tag *tag, int fid) parsedtag_get_value(tag, ATTR_METHOD, &p); q = "!CURRENT_URL!"; parsedtag_get_value(tag, ATTR_ACTION, &q); + q = url_encode(remove_space(q), cur_baseURL, cur_document_charset); r = NULL; #ifdef USE_M17N if (parsedtag_get_value(tag, ATTR_ACCEPT_CHARSET, &r)) @@ -5067,11 +5043,10 @@ HTMLtagproc1(struct parsed_tag *tag, struct html_feed_environ *h_env) } return 1; case HTML_BASE: -#ifdef USE_IMAGE +#if defined(USE_M17N) || defined(USE_IMAGE) p = NULL; if (parsedtag_get_value(tag, ATTR_HREF, &p)) { - if (!cur_baseURL) - cur_baseURL = New(ParsedURL); + cur_baseURL = New(ParsedURL); parseURL(p, cur_baseURL, NULL); } #endif @@ -5329,6 +5304,13 @@ HTMLlineproc2body(Buffer *buf, Str (*feed) (), int llimit) #ifdef MENU_SELECT Anchor **a_select = NULL; #endif +#if defined(USE_M17N) || defined(USE_IMAGE) + ParsedURL *base = baseURL(buf); +#endif +#ifdef USE_M17N + wc_ces name_charset = url_to_charset(NULL, &buf->currentURL, + buf->document_charset); +#endif if (out_size == 0) { out_size = LINELEN; @@ -5523,16 +5505,17 @@ HTMLlineproc2body(Buffer *buf, Str (*feed) (), int llimit) hseq = 0; id = NULL; if (parsedtag_get_value(tag, ATTR_NAME, &id)) { - id = url_quote_conv(id, buf->document_charset); + id = url_quote_conv(id, name_charset); registerName(buf, id, currentLn(buf), pos); } if (parsedtag_get_value(tag, ATTR_HREF, &p)) - p = url_quote_conv(remove_space(p), - buf->document_charset); + p = url_encode(remove_space(p), base, + buf->document_charset); if (parsedtag_get_value(tag, ATTR_TARGET, &q)) q = url_quote_conv(q, buf->document_charset); if (parsedtag_get_value(tag, ATTR_REFERER, &r)) - r = url_quote_conv(r, buf->document_charset); + r = url_encode(r, base, + buf->document_charset); parsedtag_get_value(tag, ATTR_TITLE, &s); parsedtag_get_value(tag, ATTR_ACCESSKEY, &t); parsedtag_get_value(tag, ATTR_HSEQ, &hseq); @@ -5618,7 +5601,7 @@ HTMLlineproc2body(Buffer *buf, Str (*feed) (), int llimit) ParsedURL u; Image *image; - parseURL2(a_img->url, &u, cur_baseURL); + parseURL2(a_img->url, &u, base); a_img->image = image = New(Image); image->url = parsedURL2Str(&u)->ptr; if (!uncompressed_file_type(u.file, &image->ext)) @@ -5639,7 +5622,7 @@ HTMLlineproc2body(Buffer *buf, Str (*feed) (), int llimit) image->map = q; image->ismap = ismap; image->touch = 0; - image->cache = getImage(image, cur_baseURL, + image->cache = getImage(image, base, IMG_FLAG_SKIP); } else if (iseq < 0) { @@ -5761,8 +5744,8 @@ HTMLlineproc2body(Buffer *buf, Str (*feed) (), int llimit) break; if (parsedtag_get_value(tag, ATTR_HREF, &p)) { MapArea *a; - p = url_quote_conv(remove_space(p), - buf->document_charset); + p = url_encode(remove_space(p), base, + buf->document_charset); t = NULL; parsedtag_get_value(tag, ATTR_TARGET, &t); q = ""; @@ -5811,11 +5794,14 @@ HTMLlineproc2body(Buffer *buf, Str (*feed) (), int llimit) break; case HTML_BASE: if (parsedtag_get_value(tag, ATTR_HREF, &p)) { - p = url_quote_conv(remove_space(p), - buf->document_charset); + p = url_encode(remove_space(p), NULL, + buf->document_charset); if (!buf->baseURL) buf->baseURL = New(ParsedURL); parseURL(p, buf->baseURL, NULL); +#if defined(USE_M17N) || defined(USE_IMAGE) + base = buf->baseURL; +#endif } if (parsedtag_get_value(tag, ATTR_TARGET, &p)) buf->baseTarget = @@ -5830,8 +5816,8 @@ HTMLlineproc2body(Buffer *buf, Str (*feed) (), int llimit) int refresh_interval = getMetaRefreshParam(q, &tmp); #ifdef USE_ALARM if (tmp) { - p = url_quote_conv(remove_space(tmp->ptr), - buf->document_charset); + p = url_encode(remove_space(tmp->ptr), base, + buf->document_charset); buf->event = setAlarmEvent(buf->event, refresh_interval, AL_IMPLICIT_ONCE, @@ -5844,8 +5830,8 @@ HTMLlineproc2body(Buffer *buf, Str (*feed) (), int llimit) FUNCNAME_reload, NULL); #else if (tmp && refresh_interval == 0) { - p = url_quote_conv(remove_space(tmp->ptr), - buf->document_charset); + p = url_encode(remove_space(tmp->ptr), base, + buf->document_charset); pushEvent(FUNCNAME_gorURL, p); } #endif @@ -5929,7 +5915,7 @@ HTMLlineproc2body(Buffer *buf, Str (*feed) (), int llimit) #ifdef ID_EXT id = NULL; if (parsedtag_get_value(tag, ATTR_ID, &id)) { - id = url_quote_conv(id, buf->document_charset); + id = url_quote_conv(id, name_charset); registerName(buf, id, currentLn(buf), pos); } if (renderFrameSet && @@ -5982,7 +5968,8 @@ addLink(Buffer *buf, struct parsed_tag *tag) parsedtag_get_value(tag, ATTR_HREF, &href); if (href) - href = url_quote_conv(remove_space(href), buf->document_charset); + href = url_encode(remove_space(href), baseURL(buf), + buf->document_charset); parsedtag_get_value(tag, ATTR_TITLE, &title); parsedtag_get_value(tag, ATTR_TYPE, &ctype); parsedtag_get_value(tag, ATTR_REL, &rel); @@ -6963,8 +6950,6 @@ loadHTMLstream(URLFile *f, Buffer *newBuf, FILE * src, int internal) image_flag = IMG_FLAG_AUTO; else image_flag = IMG_FLAG_SKIP; - if (newBuf->currentURL.file) - cur_baseURL = baseURL(newBuf); #endif if (w3m_halfload) { @@ -6987,6 +6972,9 @@ loadHTMLstream(URLFile *f, Buffer *newBuf, FILE * src, int internal) htmlenv1.f = stdout; else htmlenv1.buf = newTextLineList(); +#if defined(USE_M17N) || defined(USE_IMAGE) + cur_baseURL = baseURL(newBuf); +#endif if (SETJMP(AbortLoading) != 0) { HTMLlineproc1("
    Transfer Interrupted!
    ", &htmlenv1); @@ -7048,7 +7036,7 @@ loadHTMLstream(URLFile *f, Buffer *newBuf, FILE * src, int internal) } #endif lineBuf2 = convertLine(f, lineBuf2, HTML_MODE, &charset, doc_charset); -#if defined(USE_M17N) && defined(USE_IMAGE) +#ifdef USE_M17N cur_document_charset = charset; #endif HTMLlineproc0(lineBuf2->ptr, &htmlenv1, internal); @@ -7060,6 +7048,12 @@ loadHTMLstream(URLFile *f, Buffer *newBuf, FILE * src, int internal) obuf.status = R_ST_NORMAL; completeHTMLstream(&htmlenv1, &obuf); flushline(&htmlenv1, &obuf, 0, 2, htmlenv1.limit); +#if defined(USE_M17N) || defined(USE_IMAGE) + cur_baseURL = NULL; +#endif +#ifdef USE_M17N + cur_document_charset = 0; +#endif if (htmlenv1.title) newBuf->buffername = htmlenv1.title; if (w3m_halfdump) { @@ -7207,7 +7201,7 @@ loadGopherDir(URLFile *uf, ParsedURL *pu, wc_ces * charset) q = Strnew_m_charp("gopher://", host->ptr, ":", port->ptr, "/", file->ptr, NULL)->ptr; Strcat_m_charp(tmp, "", p, html_quote(name->ptr + 1), "\n", NULL); } @@ -7331,6 +7325,7 @@ loadImageBuffer(URLFile *uf, Buffer *newBuf) URLFile f; MySignalHandler(*volatile prevtrap) (SIGNAL_ARG) = NULL; struct stat st; + const ParsedURL *pu = newBuf ? &newBuf->currentURL : NULL; loadImage(newBuf, IMG_FLAG_STOP); image.url = uf->url; @@ -7338,8 +7333,8 @@ loadImageBuffer(URLFile *uf, Buffer *newBuf) image.width = -1; image.height = -1; image.cache = NULL; - cache = getImage(&image, cur_baseURL, IMG_FLAG_AUTO); - if (!cur_baseURL->is_nocache && cache->loaded & IMG_FLAG_LOADED && + cache = getImage(&image, (ParsedURL *)pu, IMG_FLAG_AUTO); + if (!(pu && pu->is_nocache) && cache->loaded & IMG_FLAG_LOADED && !stat(cache->file, &st)) goto image_buffer; @@ -7580,8 +7575,11 @@ openGeneralPagerBuffer(InputStream stream) #ifdef USE_M17N content_charset = 0; #endif + t_buf = newBuffer(INIT_BUFFER_WIDTH); + copyParsedURL(&t_buf->currentURL, NULL); + t_buf->currentURL.scheme = SCM_LOCAL; + t_buf->currentURL.file = "-"; if (SearchHeader) { - t_buf = newBuffer(INIT_BUFFER_WIDTH); readHeader(&uf, t_buf, TRUE, NULL); t = checkContentType(t_buf); if (t == NULL) @@ -7609,14 +7607,13 @@ openGeneralPagerBuffer(InputStream stream) #ifdef USE_IMAGE else if (activeImage && displayImage && !useExtImageViewer && !(w3m_dump & ~DUMP_FRAME) && !strncasecmp(t, "image/", 6)) { - cur_baseURL = New(ParsedURL); - parseURL("-", cur_baseURL, NULL); buf = loadImageBuffer(&uf, t_buf); buf->type = "text/html"; } #endif else { - if (doExternal(uf, "-", t, &buf, t_buf)) { + if (searchExtViewer(t)) { + buf = doExternal(uf, t, t_buf); UFclose(&uf); if (buf == NULL || buf == NO_BUFFER) return buf; @@ -7629,8 +7626,6 @@ openGeneralPagerBuffer(InputStream stream) } } buf->real_type = t; - buf->currentURL.scheme = SCM_LOCAL; - buf->currentURL.file = "-"; return buf; } @@ -7823,9 +7818,8 @@ save2tmp(URLFile uf, char *tmpf) return 0; } -int -doExternal(URLFile uf, char *path, char *type, Buffer **bufp, - Buffer *defaultbuf) +Buffer * +doExternal(URLFile uf, char *type, Buffer *defaultbuf) { Str tmpf, command; struct mailcap *mcap; @@ -7834,7 +7828,7 @@ doExternal(URLFile uf, char *path, char *type, Buffer **bufp, char *header, *src = NULL, *ext = uf.ext; if (!(mcap = searchExtViewer(type))) - return 0; + return NULL; if (mcap->nametemplate) { tmpf = unquote_mailcap(mcap->nametemplate, NULL, "", NULL, NULL); @@ -7867,15 +7861,13 @@ doExternal(URLFile uf, char *path, char *type, Buffer **bufp, UFclose(&uf); myExec(command->ptr); } - *bufp = NO_BUFFER; - return 1; + return NO_BUFFER; } else #endif { if (save2tmp(uf, tmpf->ptr) < 0) { - *bufp = NULL; - return 1; + return NULL; } } if (mcap->flags & (MAILCAP_HTMLOUTPUT | MAILCAP_COPIOUSOUTPUT)) { @@ -7918,14 +7910,13 @@ doExternal(URLFile uf, char *path, char *type, Buffer **bufp, buf = NO_BUFFER; } if (buf && buf != NO_BUFFER) { - buf->filename = path; - if (buf->buffername == NULL || buf->buffername[0] == '\0') - buf->buffername = conv_from_system(lastFileName(path)); + if ((buf->buffername == NULL || buf->buffername[0] == '\0') && + buf->filename) + buf->buffername = conv_from_system(lastFileName(buf->filename)); buf->edit = mcap->edit; buf->mailcap = mcap; } - *bufp = buf; - return 1; + return buf; } static int diff --git a/fm.h b/fm.h index 8378939..0f56c31 100644 --- a/fm.h +++ b/fm.h @@ -264,6 +264,18 @@ extern int REV_LB[]; #define IMG_FLAG_ERROR 2 #define IMG_FLAG_DONT_REMOVE 4 +#define IS_EMPTY_PARSED_URL(pu) ((pu)->scheme == SCM_UNKNOWN && !(pu)->file) +#define SCONF_RESERVED 0 +#define SCONF_SUBSTITUTE_URL 1 +#define SCONF_URL_CHARSET 2 +#define SCONF_NO_REFERER_FROM 3 +#define SCONF_NO_REFERER_TO 4 +#define SCONF_N_FIELD 5 +#define query_SCONF_SUBSTITUTE_URL(pu) ((const char *)querySiteconf(pu, SCONF_SUBSTITUTE_URL)) +#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)) + /* * Macros. */ @@ -972,6 +984,7 @@ global int BackgroundExtViewer init(TRUE); global int disable_secret_security_check init(FALSE); global char *passwd_file init(PASSWD_FILE); global char *pre_form_file init(PRE_FORM_FILE); +global char *siteconf_file init(SITECONF_FILE); global char *ftppasswd init(NULL); global int ftppass_hostnamegen init(TRUE); global int do_download init(FALSE); diff --git a/form.c b/form.c index b7556ca..fa17be4 100644 --- a/form.c +++ b/form.c @@ -787,7 +787,7 @@ struct pre_form { static struct pre_form *PreForm = NULL; static struct pre_form * -add_pre_form(struct pre_form *prev, char *url, char *name, char *action) +add_pre_form(struct pre_form *prev, char *url, Regex *re_url, char *name, char *action) { ParsedURL pu; struct pre_form *new; @@ -796,21 +796,13 @@ add_pre_form(struct pre_form *prev, char *url, char *name, char *action) new = prev->next = New(struct pre_form); else new = PreForm = New(struct pre_form); - if (url && *url == '/') { - int l = strlen(url); - if (l > 1 && url[l - 1] == '/') - new->url = allocStr(url + 1, l - 2); - else - new->url = url + 1; - new->re_url = newRegex(new->url, FALSE, NULL, NULL); - if (!new->re_url) - new->url = NULL; - } - else if (url) { + if (url && !re_url) { parseURL2(url, &pu, NULL); new->url = parsedURL2Str(&pu)->ptr; - new->re_url = NULL; } + else + new->url = url; + new->re_url = re_url; new->name = (name && *name) ? name : NULL; new->action = (action && *action) ? action : NULL; new->item = NULL; @@ -834,7 +826,7 @@ add_pre_form_item(struct pre_form *pf, struct pre_form_item *prev, int type, new->name = name; new->value = value; if (checked && *checked && (!strcmp(checked, "0") || - strcasecmp(checked, "off") + !strcasecmp(checked, "off") || !strcasecmp(checked, "no"))) new->checked = 0; else @@ -875,6 +867,7 @@ loadPreForm(void) return; while (1) { char *p, *s, *arg; + Regex *re_arg; line = Strfgets(fp); if (line->length == 0) @@ -890,18 +883,20 @@ loadPreForm(void) if (*p == '#' || *p == '\0') continue; /* comment or empty line */ s = getWord(&p); - arg = getWord(&p); if (!strcmp(s, "url")) { + arg = getRegexWord((const char **)&p, &re_arg); if (!arg || !*arg) continue; p = getQWord(&p); - pf = add_pre_form(pf, arg, NULL, p); + pf = add_pre_form(pf, arg, re_arg, NULL, p); pi = pf->item; continue; } if (!pf) continue; + + arg = getWord(&p); if (!strcmp(s, "form")) { if (!arg || !*arg) continue; @@ -913,7 +908,7 @@ loadPreForm(void) } if (pf->item) { struct pre_form *prev = pf; - pf = add_pre_form(prev, "", s, p); + pf = add_pre_form(prev, "", NULL, s, p); /* copy previous URL */ pf->url = prev->url; pf->re_url = prev->re_url; diff --git a/frame.c b/frame.c index b431437..48c2d72 100644 --- a/frame.c +++ b/frame.c @@ -91,7 +91,8 @@ newFrame(struct parsed_tag *tag, Buffer *buf) body->baseURL = baseURL(buf); if (tag) { if (parsedtag_get_value(tag, ATTR_SRC, &p)) - body->url = url_quote_conv(remove_space(p), buf->document_charset); + body->url = url_encode(remove_space(p), body->baseURL, + buf->document_charset); if (parsedtag_get_value(tag, ATTR_NAME, &p) && *p != '_') body->name = url_quote_conv(p, buf->document_charset); } @@ -639,7 +640,7 @@ createFrameFile(struct frameset *f, FILE * f1, Buffer *current, int level, case HTML_BASE: /* "BASE" is prohibit tag */ if (parsedtag_get_value(tag, ATTR_HREF, &q)) { - q = url_quote_conv(remove_space(q), charset); + q = url_encode(remove_space(q), NULL, charset); parseURL(q, &base, NULL); } if (parsedtag_get_value(tag, ATTR_TARGET, &q)) { @@ -768,8 +769,8 @@ createFrameFile(struct frameset *f, FILE * f1, Buffer *current, int level, if (!tag->value[j]) break; tag->value[j] = - url_quote_conv(remove_space(tag->value[j]), - charset); + url_encode(remove_space(tag->value[j]), + &base, charset); tag->need_reconstruct = TRUE; parseURL2(tag->value[j], &url, &base); if (url.scheme == SCM_UNKNOWN || diff --git a/func.c b/func.c index f389e00..8b5deac 100644 --- a/func.c +++ b/func.c @@ -8,6 +8,7 @@ #include "fm.h" #include "func.h" #include "myctype.h" +#include "regex.h" #include "funcname.c" #include "functable.c" @@ -434,6 +435,93 @@ getQWord(char **str) return tmp->ptr; } +/* This extracts /regex/i or m@regex@i from the given string. + * Then advances *str to the end of regex. + * If the input does not seems to be a regex, this falls back to getQWord(). + * + * Returns a word (no matter whether regex or not) in the give string. + * If regex_ret is non-NULL, compiles the regex and stores there. + * + * XXX: Actually this is unrelated to func.c. + */ +char * +getRegexWord(const char **str, Regex **regex_ret) +{ + char *word = NULL; + const char *p, *headp, *bodyp, *tailp; + char delimiter; + int esc; + int igncase = 0; + + p = *str; + SKIP_BLANKS(p); + headp = p; + + /* Get the opening delimiter */ + if (p[0] == 'm' && IS_PRINT(p[1]) && !IS_ALNUM(p[1]) && p[1] != '\\') { + delimiter = p[1]; + p += 2; + } + else if (p[0] == '/') { + delimiter = '/'; + p += 1; + } + else { + goto not_regex; + } + bodyp = p; + + /* Scan the end of the expression */ + for (esc = 0; *p; ++p) { + if (esc) { + esc = 0; + } else { + if (*p == delimiter) + break; + else if (*p == '\\') + esc = 1; + } + } + if (!*p && *headp == '/') + goto not_regex; + tailp = p; + + /* Check the modifiers */ + if (*p == delimiter) { + while (*++p && !IS_SPACE(*p)) { + switch (*p) { + case 'i': + igncase = 1; + break; + } + /* ignore unknown modifiers */ + } + } + + /* Save the expression */ + word = allocStr(headp, p - headp); + + /* Compile */ + if (regex_ret) { + if (*tailp == delimiter) + word[tailp - headp] = 0; + *regex_ret = newRegex(word + (bodyp - headp), igncase, NULL, NULL); + if (*tailp == delimiter) + word[tailp - headp] = delimiter; + } + goto last; + +not_regex: + p = headp; + word = getQWord((char **)&p); + if (regex_ret) + *regex_ret = NULL; + +last: + *str = p; + return word; +} + #ifdef USE_MOUSE static MouseAction default_mouse_action = { NULL, diff --git a/history.c b/history.c index 951ef83..e9be09b 100644 --- a/history.c +++ b/history.c @@ -17,7 +17,7 @@ historyBuffer(Hist *hist) for (item = hist->list->last; item; item = item->prev) { q = html_quote((char *)item->ptr); if (DecodeURL) - p = html_quote(url_unquote_conv((char *)item->ptr, 0)); + p = html_quote(url_decode2((char *)item->ptr, NULL)); else p = q; Strcat_charp(src, "
  • href->nanchor; i++) { ParsedURL pu; - static Str s = NULL; + char *url; if (buf->href->anchors[i].slave) continue; parseURL2(buf->href->anchors[i].url, &pu, baseURL(buf)); - s = parsedURL2Str(&pu); - if (DecodeURL) - s = Strnew_charp(url_unquote_conv - (s->ptr, Currentbuf->document_charset)); - printf("[%d] %s\n", buf->href->anchors[i].hseq + 1, s->ptr); + url = url_decode2(parsedURL2Str(&pu)->ptr, Currentbuf); + printf("[%d] %s\n", buf->href->anchors[i].hseq + 1, url); } } } @@ -2261,7 +2261,7 @@ DEFUN(movR1, MOVE_RIGHT1, static wc_uint32 getChar(char *p) { - return wc_any_to_ucs(wtf_parse1(&p)); + return wc_any_to_ucs(wtf_parse1((wc_uchar **)&p)); } static int @@ -2804,12 +2804,15 @@ loadLink(char *url, char *target, char *referer, FormList *request) union frameset_element *f_element = NULL; int flag = 0; ParsedURL *base, pu; + const int *no_referer_ptr; message(Sprintf("loading %s", url)->ptr, 0, 0); refresh(); + no_referer_ptr = query_SCONF_NO_REFERER_FROM(&Currentbuf->currentURL); base = baseURL(Currentbuf); - if (base == NULL || + if ((no_referer_ptr && *no_referer_ptr) || + base == NULL || base->scheme == SCM_LOCAL || base->scheme == SCM_LOCAL_CGI) referer = NO_REFERER; if (referer == NULL) @@ -4055,6 +4058,7 @@ goURL0(char *prompt, int relative) char *url, *referer; ParsedURL p_url, *current; Buffer *cur_buf = Currentbuf; + const int *no_referer_ptr; url = searchKeyData(); if (url == NULL) { @@ -4064,11 +4068,8 @@ goURL0(char *prompt, int relative) current = baseURL(Currentbuf); if (current) { char *c_url = parsedURL2Str(current)->ptr; - if (DefaultURLString == DEFAULT_URL_CURRENT) { - url = c_url; - if (DecodeURL) - url = url_unquote_conv(url, 0); - } + if (DefaultURLString == DEFAULT_URL_CURRENT) + url = url_decode2(c_url, NULL); else pushHist(hist, c_url); } @@ -4077,11 +4078,8 @@ goURL0(char *prompt, int relative) char *a_url; parseURL2(a->url, &p_url, current); a_url = parsedURL2Str(&p_url)->ptr; - if (DefaultURLString == DEFAULT_URL_LINK) { - url = a_url; - if (DecodeURL) - url = url_unquote_conv(url, Currentbuf->document_charset); - } + if (DefaultURLString == DEFAULT_URL_LINK) + url = url_decode2(a_url, Currentbuf); else pushHist(hist, a_url); } @@ -4089,15 +4087,22 @@ goURL0(char *prompt, int relative) if (url != NULL) SKIP_BLANKS(url); } -#ifdef USE_M17N - if (url != NULL) { - if ((relative || *url == '#') && Currentbuf->document_charset) - url = wc_conv_strict(url, InnerCharset, - Currentbuf->document_charset)->ptr; + if (relative) { + no_referer_ptr = query_SCONF_NO_REFERER_FROM(&Currentbuf->currentURL); + current = baseURL(Currentbuf); + if ((no_referer_ptr && *no_referer_ptr) || + current == NULL || + current->scheme == SCM_LOCAL || current->scheme == SCM_LOCAL_CGI) + referer = NO_REFERER; else - url = conv_to_system(url); + referer = parsedURL2Str(&Currentbuf->currentURL)->ptr; + url = url_encode(url, current, Currentbuf->document_charset); + } + else { + current = NULL; + referer = NULL; + url = url_encode(url, NULL, 0); } -#endif if (url == NULL || *url == '\0') { displayBuffer(Currentbuf, B_FORCE_REDRAW); return; @@ -4106,14 +4111,6 @@ goURL0(char *prompt, int relative) gotoLabel(url + 1); return; } - if (relative) { - current = baseURL(Currentbuf); - referer = parsedURL2Str(&Currentbuf->currentURL)->ptr; - } - else { - current = NULL; - referer = NULL; - } parseURL2(url, &p_url, current); pushHashHist(URLHist, parsedURL2Str(&p_url)->ptr); cmd_loadURL(url, current, referer, NULL); @@ -4510,8 +4507,7 @@ _peekURL(int only_img) s = parsedURL2Str(&pu); } if (DecodeURL) - s = Strnew_charp(url_unquote_conv - (s->ptr, Currentbuf->document_charset)); + s = Strnew_charp(url_decode2(s->ptr, Currentbuf)); #ifdef USE_M17N s = checkType(s, &pp, NULL); p = NewAtom_N(Lineprop, s->length); @@ -4570,7 +4566,7 @@ DEFUN(curURL, PEEK, "Peek current URL") offset = 0; s = currentURL(); if (DecodeURL) - s = Strnew_charp(url_unquote_conv(s->ptr, 0)); + s = Strnew_charp(url_decode2(s->ptr, NULL)); #ifdef USE_M17N s = checkType(s, &pp, NULL); p = NewAtom_N(Lineprop, s->length); diff --git a/map.c b/map.c index 90aa35a..12701e7 100644 --- a/map.c +++ b/map.c @@ -279,7 +279,7 @@ follow_map_panel(Buffer *buf, char *name) p = parsedURL2Str(&pu)->ptr; q = html_quote(p); if (DecodeURL) - p = html_quote(url_unquote_conv(p, buf->document_charset)); + p = html_quote(url_decode2(p, buf)); else p = q; Strcat_m_charp(mappage, "", @@ -417,10 +417,7 @@ append_map_info(Buffer *buf, Str tmp, FormItemList *fi) continue; parseURL2(a->url, &pu, baseURL(buf)); q = html_quote(parsedURL2Str(&pu)->ptr); - if (DecodeURL) - p = html_quote(url_unquote_conv(a->url, buf->document_charset)); - else - p = html_quote(a->url); + p = html_quote(url_decode2(a->url, buf)); Strcat_m_charp(tmp, "  ", html_quote(*a->alt ? a->alt : mybasename(a->url)), @@ -457,10 +454,8 @@ append_link_info(Buffer *buf, Str html, LinkList * link) Strcat_charp(html, "[Rev]"); if (!l->url) url = "(empty)"; - else if (DecodeURL) - url = html_quote(url_unquote_conv(l->url, buf->document_charset)); else - url = html_quote(l->url); + url = html_quote(url_decode2(l->url, buf)); Strcat_m_charp(html, "", url, NULL); if (l->ctype) Strcat_m_charp(html, " (", html_quote(l->ctype), ")", NULL); @@ -498,8 +493,7 @@ append_frame_info(Buffer *buf, Str html, struct frameset *set, int level) Strcat_charp(html, p); } if (DecodeURL) - p = html_quote(url_unquote_conv(frame.body->url, - buf->document_charset)); + p = html_quote(url_decode2(frame.body->url, buf)); else p = q; Strcat_m_charp(html, " ", p, "
    \n", NULL); @@ -550,9 +544,7 @@ page_info_panel(Buffer *buf) #ifdef USE_M17N Strcat_charp(tmp, "
    "); #endif - p = parsedURL2Str(&buf->currentURL)->ptr; - if (DecodeURL) - p = url_unquote_conv(p, 0); + p = url_decode2(parsedURL2Str(&buf->currentURL)->ptr, NULL); Strcat_m_charp(tmp, "", "
    Title", html_quote(buf->buffername), @@ -589,7 +581,7 @@ page_info_panel(Buffer *buf) p = parsedURL2Str(&pu)->ptr; q = html_quote(p); if (DecodeURL) - p = html_quote(url_unquote_conv(p, buf->document_charset)); + p = html_quote(url_decode2(p, buf)); else p = q; Strcat_m_charp(tmp, @@ -602,7 +594,7 @@ page_info_panel(Buffer *buf) p = parsedURL2Str(&pu)->ptr; q = html_quote(p); if (DecodeURL) - p = html_quote(url_unquote_conv(p, buf->document_charset)); + p = html_quote(url_decode2(p, buf)); else p = q; Strcat_m_charp(tmp, @@ -613,10 +605,7 @@ page_info_panel(Buffer *buf) if (a != NULL) { FormItemList *fi = (FormItemList *)a->url; p = form2str(fi); - if (DecodeURL) - p = html_quote(url_unquote_conv(p, buf->document_charset)); - else - p = html_quote(p); + p = html_quote(url_decode2(p, buf)); Strcat_m_charp(tmp, "
    Method/type of current form ", p, NULL); diff --git a/menu.c b/menu.c index 774b1bd..0f66583 100644 --- a/menu.c +++ b/menu.c @@ -1365,9 +1365,7 @@ initSelectMenu(void) break; default: Strcat_char(str, ' '); - p = parsedURL2Str(&buf->currentURL)->ptr; - if (DecodeURL) - p = url_unquote_conv(p, 0); + p = url_decode2(parsedURL2Str(&buf->currentURL)->ptr, NULL); Strcat_charp(str, p); break; } @@ -1513,9 +1511,7 @@ initSelTabMenu(void) case SCM_MISSING: break; default: - p = parsedURL2Str(&buf->currentURL)->ptr; - if (DecodeURL) - p = url_unquote_conv(p, 0); + p = url_decode2(parsedURL2Str(&buf->currentURL)->ptr, NULL); Strcat_charp(str, p); break; } @@ -1845,10 +1841,8 @@ link_menu(Buffer *buf) Strcat_charp(str, " "); if (!l->url) p = ""; - else if (DecodeURL) - p = url_unquote_conv(l->url, buf->document_charset); else - p = l->url; + p = url_decode2(l->url, buf); Strcat_charp(str, p); label[i] = str->ptr; if (len < str->length) diff --git a/po/ja.po b/po/ja.po index d67c695..947191c 100644 --- a/po/ja.po +++ b/po/ja.po @@ -407,6 +407,10 @@ msgid "File for setting form on loading" msgstr "梧莨惹若荐絎＜ゃ" #: rc.c:149 +msgid "File for preferences for each site" +msgstr "泣ゃヨ┃絎＜ゃ" + +#: rc.c:149 msgid "Password for anonymous FTP (your mail address)" msgstr "FTP鴻若(mail address篏帥)" diff --git a/proto.h b/proto.h index f8a7345..7b1a7a6 100644 --- a/proto.h +++ b/proto.h @@ -162,6 +162,24 @@ extern Str searchURIMethods(ParsedURL *pu); extern void chkExternalURIBuffer(Buffer *buf); #endif extern ParsedURL *schemeToProxy(int scheme); +#ifdef USE_M17N +extern wc_ces url_to_charset(const char *url, const ParsedURL *base, + wc_ces doc_charset); +extern char *url_encode(const char *url, const ParsedURL *base, + wc_ces doc_charset); +#if 0 +extern char *url_decode(const char *url, const ParsedURL *base, + wc_ces doc_charset); +#endif +extern char *url_decode2(const char *url, const Buffer *buf); +#else /* !defined(USE_M17N) */ +#define url_encode(url, base, cs) url_quote(url) +extern char *url_decode0(const char *url); +#if 0 +#define url_decode(url, base, cs) url_decode0(url) +#endif +#define url_decode2(url, buf) url_decode0(url) +#endif /* !defined(USE_M17N) */ extern void examineFile(char *path, URLFile *uf); extern char *acceptableEncoding(); extern int dir_exist(char *path); @@ -180,7 +198,6 @@ extern void push_symbol(Str str, char symbol, int width, int n); #ifdef USE_UNICODE extern void update_utf8_symbol(void); #endif -extern Buffer *loadFile(char *path); extern Buffer *loadGeneralFile(char *path, ParsedURL *current, char *referer, int flag, FormList *request); extern int is_boundary(unsigned char *, unsigned char *); @@ -249,8 +266,7 @@ extern Buffer *openPagerBuffer(InputStream stream, Buffer *buf); extern Buffer *openGeneralPagerBuffer(InputStream stream); extern Line *getNextPage(Buffer *buf, int plen); extern int save2tmp(URLFile uf, char *tmpf); -extern int doExternal(URLFile uf, char *path, char *type, Buffer **bufp, - Buffer *defaultbuf); +extern Buffer *doExternal(URLFile uf, char *type, Buffer *defaultbuf); extern int _doFileCopy(char *tmpf, char *defstr, int download); #define doFileCopy(tmpf, defstr) _doFileCopy(tmpf, defstr, FALSE); extern int doFileMove(char *tmpf, char *defstr); @@ -507,7 +523,7 @@ extern ParsedURL *baseURL(Buffer *buf); extern int openSocket(char *hostname, char *remoteport_name, unsigned short remoteport_num); extern void parseURL(char *url, ParsedURL *p_url, ParsedURL *current); -extern void copyParsedURL(ParsedURL *p, ParsedURL *q); +extern void copyParsedURL(ParsedURL *p, const ParsedURL *q); extern void parseURL2(char *url, ParsedURL *pu, ParsedURL *current); extern Str parsedURL2Str(ParsedURL *pu); extern int getURLScheme(char **url); @@ -611,6 +627,7 @@ extern char *confFile(char *base); extern char *auxbinFile(char *base); extern char *libFile(char *base); extern char *helpFile(char *base); +extern const void *querySiteconf(const ParsedURL *query_pu, int field); extern Str localCookie(void); extern Str loadLocalDir(char *dirname); extern void set_environ(char *var, char *value); @@ -723,6 +740,8 @@ extern int getKey(char *s); extern char *getKeyData(int key); extern char *getWord(char **str); extern char *getQWord(char **str); +struct regex; +extern char *getRegexWord(const char **str, struct regex **regex_ret); #ifdef USE_MOUSE extern void initMouseAction(void); #endif diff --git a/rc.c b/rc.c index 8441a39..3bf6cea 100644 --- a/rc.c +++ b/rc.c @@ -9,7 +9,9 @@ #include #include "parsetag.h" #include "local.h" +#include "regex.h" #include +#include struct param_ptr { char *name; @@ -146,6 +148,7 @@ static int OptionEncode = FALSE; #define CMT_DISABLE_SECRET_SECURITY_CHECK N_("Disable secret file security check") #define CMT_PASSWDFILE N_("Password file") #define CMT_PRE_FORM_FILE N_("File for setting form on loading") +#define CMT_SITECONF_FILE N_("File for preferences for each site") #define CMT_FTPPASS N_("Password for anonymous FTP (your mail address)") #define CMT_FTPPASS_HOSTNAMEGEN N_("Generate domain part of password for FTP") #define CMT_USERAGENT N_("User-Agent identification string") @@ -619,6 +622,8 @@ struct param_ptr params9[] = { CMT_FTPPASS_HOSTNAMEGEN, NULL}, {"pre_form_file", P_STRING, PI_TEXT, (void *)&pre_form_file, CMT_PRE_FORM_FILE, NULL}, + {"siteconf_file", P_STRING, PI_TEXT, (void *)&siteconf_file, + CMT_SITECONF_FILE, NULL}, {"user_agent", P_STRING, PI_TEXT, (void *)&UserAgent, CMT_USERAGENT, NULL}, {"no_referer", P_INT, PI_ONOFF, (void *)&NoSendReferer, CMT_NOSENDREFERER, NULL}, @@ -1173,6 +1178,8 @@ do_mkdir(const char *dir, long mode) #endif /* not __MINW32_VERSION */ #endif /* not __EMX__ */ +static void loadSiteconf(void); + void sync_with_option(void) { @@ -1199,6 +1206,7 @@ sync_with_option(void) #endif loadPasswd(); loadPreForm(); + loadSiteconf(); if (AcceptLang == NULL || *AcceptLang == '\0') { /* TRANSLATORS: @@ -1556,3 +1564,217 @@ helpFile(char *base) return expandPath(Strnew_m_charp(w3m_help_dir(), "/", base, NULL)->ptr); } #endif + +/* siteconf */ +/* + * url ""|//|m@@i [exact] + * substitute_url "" + * url_charset + * no_referer_from on|off + * no_referer_to on|off + * + * The last match wins. + */ + +struct siteconf_rec { + struct siteconf_rec *next; + char *url; + Regex *re_url; + int url_exact; + unsigned char mask[(SCONF_N_FIELD + 7) >> 3]; + + char *substitute_url; +#ifdef USE_M17N + wc_ces url_charset; +#endif + int no_referer_from; + int no_referer_to; +}; +#define SCONF_TEST(ent, f) ((ent)->mask[(f)>>3] & (1U<<((f)&7))) +#define SCONF_SET(ent, f) ((ent)->mask[(f)>>3] |= (1U<<((f)&7))) +#define SCONF_CLEAR(ent, f) ((ent)->mask[(f)>>3] &= ~(1U<<((f)&7))) + +static struct siteconf_rec *siteconf_head = NULL; +static struct siteconf_rec *newSiteconfRec(void); + +static struct siteconf_rec * +newSiteconfRec(void) +{ + struct siteconf_rec *ent; + + ent = New(struct siteconf_rec); + ent->next = NULL; + ent->url = NULL; + ent->re_url = NULL; + ent->url_exact = FALSE; + memset(ent->mask, 0, sizeof(ent->mask)); + + ent->substitute_url = NULL; +#ifdef USE_M17N + ent->url_charset = 0; +#endif + return ent; +} + +static void +loadSiteconf(void) +{ + char *efname; + FILE *fp; + Str line; + struct siteconf_rec *ent = NULL; + + siteconf_head = NULL; + if (!siteconf_file) + return; + if ((efname = expandPath(siteconf_file)) == NULL) + return; + fp = fopen(efname, "r"); + if (fp == NULL) + return; + while (line = Strfgets(fp), line->length > 0) { + char *p, *s; + + Strchop(line); + p = line->ptr; + SKIP_BLANKS(p); + if (*p == '#' || *p == '\0') + continue; + s = getWord(&p); + + /* The "url" begins a new record. */ + if (strcmp(s, "url") == 0) { + char *url, *opt; + struct siteconf_rec *newent; + + /* First, register the current record. */ + if (ent) { + ent->next = siteconf_head; + siteconf_head = ent; + ent = NULL; + } + + /* Second, create a new record. */ + newent = newSiteconfRec(); + url = getRegexWord((const char **)&p, &newent->re_url); + opt = getWord(&p); + SKIP_BLANKS(p); + if (!newent->re_url) { + ParsedURL pu; + if (!url || !*url) + continue; + parseURL2(url, &pu, NULL); + newent->url = parsedURL2Str(&pu)->ptr; + } + /* If we have an extra or unknown option, ignore this record + * for future extensions. */ + if (strcmp(opt, "exact") == 0) { + newent->url_exact = TRUE; + } + else if (*opt != 0) + continue; + if (*p) + continue; + ent = newent; + continue; + } + + /* If the current record is broken, skip to the next "url". */ + if (!ent) + continue; + + /* Fill the new record. */ + if (strcmp(s, "substitute_url") == 0) { + ent->substitute_url = getQWord(&p); + SCONF_SET(ent, SCONF_SUBSTITUTE_URL); + } +#ifdef USE_M17N + else if (strcmp(s, "url_charset") == 0) { + char *charset = getWord(&p); + ent->url_charset = (charset && *charset) ? + wc_charset_to_ces(charset) : 0; + SCONF_SET(ent, SCONF_URL_CHARSET); + } +#endif /* USE_M17N */ + else if (strcmp(s, "no_referer_from") == 0) { + ent->no_referer_from = str_to_bool(getWord(&p), 0); + SCONF_SET(ent, SCONF_NO_REFERER_FROM); + } + else if (strcmp(s, "no_referer_to") == 0) { + ent->no_referer_to = str_to_bool(getWord(&p), 0); + SCONF_SET(ent, SCONF_NO_REFERER_TO); + } + } + if (ent) { + ent->next = siteconf_head; + siteconf_head = ent; + ent = NULL; + } + fclose(fp); +} + +const void * +querySiteconf(const ParsedURL *query_pu, int field) +{ + const struct siteconf_rec *ent; + Str u; + char *firstp, *lastp; + + if (field < 0 || field >= SCONF_N_FIELD) + return NULL; + if (!query_pu || IS_EMPTY_PARSED_URL(query_pu)) + return NULL; + u = parsedURL2Str((ParsedURL *)query_pu); + if (u->length == 0) + return NULL; + + for (ent = siteconf_head; ent; ent = ent->next) { + if (!SCONF_TEST(ent, field)) + continue; + if (ent->re_url) { + if (RegexMatch(ent->re_url, u->ptr, u->length, 1)) { + MatchedPosition(ent->re_url, &firstp, &lastp); + if (!ent->url_exact) + goto url_found; + if (firstp != u->ptr || lastp == firstp) + continue; + if (*lastp == 0 || *lastp == '?' || *(lastp - 1) == '?' || + *lastp == '#' || *(lastp - 1) == '#') + goto url_found; + } + } else { + int matchlen = strmatchlen(ent->url, u->ptr, u->length); + if (matchlen == 0 || ent->url[matchlen] != 0) + continue; + firstp = u->ptr; + lastp = u->ptr + matchlen; + if (*lastp == 0 || *lastp == '?' || *(lastp - 1) == '?' || + *lastp == '#' || *(lastp - 1) == '#') + goto url_found; + if (!ent->url_exact && (*lastp == '/' || *(lastp - 1) == '/')) + goto url_found; + } + } + return NULL; + +url_found: + switch (field) { + case SCONF_SUBSTITUTE_URL: + if (ent->substitute_url && *ent->substitute_url) { + Str tmp = Strnew_charp_n(u->ptr, firstp - u->ptr); + Strcat_charp(tmp, ent->substitute_url); + Strcat_charp(tmp, lastp); + return tmp->ptr; + } + return NULL; +#ifdef USE_M17N + case SCONF_URL_CHARSET: + return &ent->url_charset; +#endif + case SCONF_NO_REFERER_FROM: + return &ent->no_referer_from; + case SCONF_NO_REFERER_TO: + return &ent->no_referer_to; + } + return NULL; +} diff --git a/url.c b/url.c index ed6062e..cbb4aab 100644 --- a/url.c +++ b/url.c @@ -444,6 +444,8 @@ baseURL(Buffer *buf) /* tag is defined in the document */ return buf->baseURL; } + else if (IS_EMPTY_PARSED_URL(&buf->currentURL)) + return NULL; else return &buf->currentURL; } @@ -638,16 +640,21 @@ openSocket(char *const hostname, #define COPYPATH_SPC_ALLOW 0 #define COPYPATH_SPC_IGNORE 1 #define COPYPATH_SPC_REPLACE 2 +#define COPYPATH_SPC_MASK 3 +#define COPYPATH_LOWERCASE 4 static char * copyPath(char *orgpath, int length, int option) { Str tmp = Strnew(); - while (*orgpath && length != 0) { - if (IS_SPACE(*orgpath)) { - switch (option) { + char ch; + while ((ch = *orgpath) != 0 && length != 0) { + if (option & COPYPATH_LOWERCASE) + ch = TOLOWER(ch); + if (IS_SPACE(ch)) { + switch (option & COPYPATH_SPC_MASK) { case COPYPATH_SPC_ALLOW: - Strcat_char(tmp, *orgpath); + Strcat_char(tmp, ch); break; case COPYPATH_SPC_IGNORE: /* do nothing */ @@ -658,7 +665,7 @@ copyPath(char *orgpath, int length, int option) } } else - Strcat_char(tmp, *orgpath); + Strcat_char(tmp, ch); orgpath++; length--; } @@ -668,22 +675,14 @@ copyPath(char *orgpath, int length, int option) void parseURL(char *url, ParsedURL *p_url, ParsedURL *current) { - char *p, *q; + char *p, *q, *qq; Str tmp; url = url_quote(url); /* quote 0x01-0x20, 0x7F-0xFF */ p = url; + copyParsedURL(p_url, NULL); p_url->scheme = SCM_MISSING; - p_url->port = 0; - p_url->user = NULL; - p_url->pass = NULL; - p_url->host = NULL; - p_url->is_nocache = 0; - p_url->file = NULL; - p_url->real_file = NULL; - p_url->query = NULL; - p_url->label = NULL; /* RFC1808: Relative Uniform Resource Locators * 4. Resolving Relative URLs @@ -694,7 +693,7 @@ parseURL(char *url, ParsedURL *p_url, ParsedURL *current) goto do_label; } #if defined( __EMX__ ) || defined( __CYGWIN__ ) - if (!strncmp(url, "file://localhost/", 17)) { + if (!strncasecmp(url, "file://localhost/", 17)) { p_url->scheme = SCM_LOCAL; p += 17 - 1; url += 17 - 1; @@ -802,19 +801,20 @@ parseURL(char *url, ParsedURL *p_url, ParsedURL *current) /* scheme://user:pass@host or * scheme://host:port */ - p_url->host = copyPath(q, p - q, COPYPATH_SPC_IGNORE); + qq = q; q = ++p; while (*p && strchr("@/?#", *p) == NULL) p++; if (*p == '@') { /* scheme://user:pass@... */ + p_url->user = copyPath(qq, q - 1 - qq, COPYPATH_SPC_IGNORE); p_url->pass = copyPath(q, p - q, COPYPATH_SPC_ALLOW); q = ++p; - p_url->user = p_url->host; - p_url->host = NULL; goto analyze_url; } /* scheme://host:port/ */ + p_url->host = copyPath(qq, q - 1 - qq, + COPYPATH_SPC_IGNORE | COPYPATH_LOWERCASE); tmp = Strnew_charp_n(q, p - q); p_url->port = atoi(tmp->ptr); /* *p is one of ['\0', '/', '?', '#'] */ @@ -829,7 +829,8 @@ parseURL(char *url, ParsedURL *p_url, ParsedURL *current) case '/': case '?': case '#': - p_url->host = copyPath(q, p - q, COPYPATH_SPC_IGNORE); + p_url->host = copyPath(q, p - q, + COPYPATH_SPC_IGNORE | COPYPATH_LOWERCASE); p_url->port = DefaultPort[p_url->scheme]; break; } @@ -956,12 +957,16 @@ parseURL(char *url, ParsedURL *p_url, ParsedURL *current) p_url->label = NULL; } -#define initParsedURL(p) bzero(p,sizeof(ParsedURL)) #define ALLOC_STR(s) ((s)==NULL?NULL:allocStr(s,-1)) void -copyParsedURL(ParsedURL *p, ParsedURL *q) +copyParsedURL(ParsedURL *p, const ParsedURL *q) { + if (q == NULL) { + memset(p, 0, sizeof(ParsedURL)); + p->scheme = SCM_UNKNOWN; + return; + } p->scheme = q->scheme; p->port = q->port; p->is_nocache = q->is_nocache; @@ -1283,6 +1288,8 @@ static char * otherinfo(ParsedURL *target, ParsedURL *current, char *referer) { Str s = Strnew(); + const int *no_referer_ptr; + int no_referer; Strcat_charp(s, "User-Agent: "); if (UserAgent == NULL || *UserAgent == '\0') @@ -1306,7 +1313,12 @@ otherinfo(ParsedURL *target, ParsedURL *current, char *referer) Strcat_charp(s, "Pragma: no-cache\r\n"); Strcat_charp(s, "Cache-control: no-cache\r\n"); } - if (!NoSendReferer) { + no_referer = NoSendReferer; + no_referer_ptr = query_SCONF_NO_REFERER_FROM(current); + no_referer = NoSendReferer || (no_referer_ptr && *no_referer_ptr); + no_referer_ptr = query_SCONF_NO_REFERER_TO(target); + no_referer = no_referer || (no_referer_ptr && *no_referer_ptr); + if (!no_referer) { #ifdef USE_SSL if (current && current->scheme == SCM_HTTPS && target->scheme != SCM_HTTPS) { /* Don't send Referer: if https:// -> http:// */ @@ -1314,6 +1326,7 @@ otherinfo(ParsedURL *target, ParsedURL *current, char *referer) else #endif if (referer == NULL && current && current->scheme != SCM_LOCAL && + current->scheme != SCM_LOCAL_CGI && (current->scheme != SCM_FTP || (current->user == NULL && current->pass == NULL))) { char *p = current->label; @@ -2234,3 +2247,66 @@ schemeToProxy(int scheme) } return pu; } + +#ifdef USE_M17N +wc_ces +url_to_charset(const char *url, const ParsedURL *base, wc_ces doc_charset) +{ + const ParsedURL *pu; + ParsedURL pu_buf; + const wc_ces *csptr; + + if (url && *url && *url != '#') { + parseURL2((char *)url, &pu_buf, (ParsedURL *)base); + pu = &pu_buf; + } else { + pu = base; + } + if (pu && (pu->scheme == SCM_LOCAL || pu->scheme == SCM_LOCAL_CGI)) + return SystemCharset; + csptr = query_SCONF_URL_CHARSET(pu); + return (csptr && *csptr) ? *csptr : + doc_charset ? doc_charset : DocumentCharset; +} + +char * +url_encode(const char *url, const ParsedURL *base, wc_ces doc_charset) +{ + return url_quote_conv((char *)url, + url_to_charset(url, base, doc_charset)); +} + +#if 0 /* unused */ +char * +url_decode(const char *url, const ParsedURL *base, wc_ces doc_charset) +{ + if (!DecodeURL) + return (char *)url; + return url_unquote_conv((char *)url, + url_to_charset(url, base, doc_charset)); +} +#endif + +char * +url_decode2(const char *url, const Buffer *buf) +{ + wc_ces url_charset; + + if (!DecodeURL) + return (char *)url; + url_charset = buf ? + url_to_charset(url, baseURL((Buffer *)buf), buf->document_charset) : + url_to_charset(url, NULL, 0); + return url_unquote_conv((char *)url, url_charset); +} + +#else /* !defined(USE_M17N) */ + +char * +url_decode0(const char *url) +{ + if (!DecodeURL) + return (char *)url; + return url_unquote_conv((char *)url, 0); +} +#endif /* !defined(USE_M17N) */